YT YouTube2Text

YouTube2Text → Transcripts → Probability Concepts (2025 CFA® Level I Exam – Quantitative Methods – Module 3)

Probability Concepts (2025 CFA® Level I Exam – Quantitative Methods – Module 3) — Transcript

by AnalystPrep · 9,919 words · 1,406 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:04this is level one of the CFA program the
0:06topic on quantitative methods and the
0:09reading on probability Concepts from my
0:12perspective this is among the most
0:14important readings in level one because
0:17it provides us with a great Foundation
0:20to apply these probability Concepts to
0:23other financial decisions that we're
0:25going to make throughout level one and
0:27then of course level two and level three
0:30we're surrounded by probabilities right
0:32every day we ask the question what's the
0:34probability is it going to rain or slow
0:36or snow what's the probability that the
0:38train will be on time what's the
0:40probability my
0:41car um will take me to work
0:44probabilities probabilities right but as
0:46good Financial analysts we want to know
0:48the answer to these questions what's the
0:50probability of a stock increasing by the
0:52end of the year what's the probability
0:55of a bond defaulting by the end of the
0:57year what's the probability of an option
0:59finishing in the money and you'll see
1:02this kind of a sense of this general
1:04idea of probabilities and probability
1:07Concepts as they can be applied to
1:10specific
1:12events and that's really that first Los
1:15there but I want you to skip down to
1:16look at the third and the fourth
1:18describe probability of an event
1:20calculate interpret conditional
1:22probabilities so all of the other loss's
1:25are pretty much surrounding those two we
1:28want to describe and calculate different
1:31types of probabilities now of course
1:34we're going to need to know some
1:35supporting information and have a good
1:38understanding of things like Mutual
1:41exclusivity empirical or subjective and
1:44a priori probabilities dependent and
1:48independent events events we're going to
1:50have to know some probability
1:53rules and then as we begin the second
1:56page of the Los is we're going to
1:59advance over to portfolio Theory when we
2:02talk about expected values and standard
2:04deviations we'll look at probability
2:06trees we'll go ahead and look at uh
2:09briefly uh bay Theory and then we'll end
2:13with some uh counting issues that you
2:16probably learned back in you know fifth
2:18or seventh grade whatever whatever time
2:20period you learned about factoral so
2:22let's go ahead and start with this uh
2:24relatively straightforward definition in
2:26the first Los so a random variable
2:29refers to any quantity with uncertain
2:32expected future
2:34values now this is what I was saying
2:36earlier in the introduction rate of
2:38return earned on a stock next year value
2:40of some kind of an index or or the time
2:43of death So Random variables
2:46remember and I always think of a random
2:49variable using the Excel function uh the
2:53random variable function you know if you
2:54hit that thing and and you hit the enter
2:57any number can show up in that
2:59particular particular cell that's what a
3:01random variable is an outcome refers to
3:04any possible value that a random
3:06variable can take right time of death
3:09could be you know
3:122040 uh for me could be uh S&P 500
3:17increases by 22% uh rate of return on a
3:20stock could be 16% or minus
3:2422% and then event is a specified
3:27outcome or a specified set of
3:33outcomes let's continue on with this uh
3:36next losos which really is uh let's look
3:39at some more definitions so what does
3:41the term probability mean a number
3:44between zero and one and Skip down to
3:47that third bullet point here a certain
3:48event has a probability of 100% this is
3:52going to be really important when we
3:54talk about this concept of a risk-free
3:56rate of return going forward to finish
3:59level one and then of course level two
4:01and level
4:02three uh I'm guessing you know that we
4:04use uh US Treasury Securities as a proxy
4:08for that risk-free rate of return well
4:11what that means is that US federal
4:12government with 100% certainty is going
4:15to make those promised payments now that
4:18might be the only case where we have a
4:20100% probability when we talk about uh
4:25uh financial securities and financial
4:27markets although when we talk about der
4:29Rives and I show you some examples there
4:32there there are opportunities out there
4:34to perfectly hedge which means you'll
4:36have a 100% probability but that's
4:38that's a conversation for a future day
4:41if an event is impossible then its
4:43probability is
4:45zero uh the probability that my children
4:49collectively uh will enjoy tonight's
4:52prepared dinner that will be
4:560% so between between zero and one means
4:59that 10% 20% 50% 80% so so we can define
5:04those probabilities between zero and
5:06one so the two defining probabilities
5:10any number between zero and one and
5:13inclusive and then of course the sum of
5:14those
5:15probabilities uh must equal one because
5:18something's going to happen right if you
5:20have a 20% chance of rain and an 80%
5:23chance of sunshine well there's 100%
5:25chance of something happening today up
5:27in the atmosphere
5:31how about mutually exclusive events it
5:34means that only one of them can take
5:37place at any time my my example that I
5:40give students all the time and this
5:42might resonate with you is that the
5:43marital decision and the marriage event
5:46is a mutual exclusive event when you say
5:50I do to that one individual you're
5:52saying I don't to all of those other
5:54people out there so event a return on
5:56the stock equals 10% event B the return
5:59on the stock is less than 10% and we use
6:02that example so that you can understand
6:05that uh there is a definition in which
6:08events can cover all possible outcomes
6:10and this is the term we use is
6:13exhaustive and so look at this example
6:16uh down below event a return is equal to
6:1910 event B return is less than 10 and
6:23then return uh event C return is greater
6:26than 10
6:28so uh
6:30that example at the bottom covers all
6:32possible
6:33outcomes now different types of
6:36probabilities and empirical probability
6:39results from analyzing actual historical
6:43data and what you're going to do and
6:45we'll do this pretty regularly
6:47throughout levels one two and three is
6:48we're going to take a look at some
6:50historical returns and use that as a
6:52basis to estimate what's going to happen
6:55during the course of next year of course
6:58in my personal example could say
7:00something like none of my four children
7:02have ever given a thumbs up to any
7:04prepared meal that my wife and I have
7:07made over the last year so given that
7:11that the empirical probability of
7:12tonight's reception is going to be uh
7:15probably the same as it always has
7:17been now of course there's a little bit
7:20of subjectivity in my personal example
7:22there but let's make sure you can
7:23separate that so subjective
7:25probabilities reflect personal beliefs
7:28or judgment personal experience based uh
7:30so that you say something like all right
7:33here we go you ready this is going to
7:35introduce the element of behavior whole
7:38Finance you know oh I love this stock
7:40let's just suppose it's Amazon stock
7:42because it's been going up and up and up
7:43for the last you know however many years
7:45and so I love it for the next uh 20
7:47years because uh my wife and I order
7:50something from Amazon two or three times
7:52a week and the Amazon delivery man or
7:55woman is so happy and they're so
7:57delighted when they come to our front
7:58door and I get this gift it's like
8:01Christmas several times a week so I'm
8:03using my personal beliefs to kind of say
8:05something about the return on an
8:08individual stock that may or may not
8:10have any foundation in uh economic
8:15support a priori probabilities are
8:18deductive and based on reasoning which
8:22means we can take a look at the data and
8:23we can make some kind of a let me go
8:26ahead and say it some kind of a
8:27mathematical maybe Quant ative is a
8:30better word uh
8:32decision so there's an example two
8:34applicants candidate a 70% chance then
8:38we can look and say hey you know what
8:40candidate B only has a 30% chance
8:43reasoning there that makes
8:46sense all right how about odds for and
8:49against uh an event so this is just
8:52going to be a ratio and so look at the
8:55uh look at the second um the second
8:58arrow point they're given odds for
9:01E of a to B the implied probability is
9:05just a ratio of a over a plus b and so
9:09then the same is true for against an
9:11event and let's go ahead and do just a
9:13quick example here this should make
9:14perfect
9:15sense uh an analyst believes that uh
9:18earnings per share will range from uh 20
9:22to 35 cents next year uh the odds for
9:25the stock beating that highest estimate
9:28uh 35 cents are 1 to n so that tells us
9:32one to n that's probably 10 outcomes so
9:35what is the probability that the stock
9:36will actually beat the highest estimate
9:40so there's the one in the numerator and
9:42then the 1+ 9 in the denominator so so
9:45that's
9:4710% how about a conditional probability
9:50oh now we're you know back here we just
9:53had some information just basic
9:55information but now let's suppose that
9:56we have some extra information and
9:59that's what a conditional probability
10:01does for so it describes the probability
10:03of an event given that something else
10:06has already occurred you know let's
10:09suppose that we're uh trying to predict
10:12a conditional probability or just the
10:15probability that a particular football
10:17team is going to win a game this
10:19Saturday if it's college or this Sunday
10:21if it's professional and so what we can
10:23do is we can look at the history history
10:26and we can you know kind of give some
10:27kind of a general probability which is
10:29what you know we've done up to this
10:31point in the recording but let's suppose
10:34that uh these two teams are evenly
10:36matched and
10:38that um over the last 10 years each team
10:42has won five times so you probably would
10:44think hey you know what that uh there's
10:48probably a 50% chance that either one of
10:50these teams is going to win this coming
10:53weekend but suppose suppose that the
10:56five times that team a won it snowed or
11:00rained or had some kind of weather event
11:02and the five times that Team B won it
11:04was sunny and it was beautiful so
11:06wouldn't it be relevant wouldn't it be
11:09critical to our estimate of a
11:11probability of this coming weekend's
11:14outcome if we knew what the weather was
11:16going to be right if we knew the weather
11:19then we would be able to predict much
11:22more accurately what that outcome would
11:25be so here's some notation we're going
11:28to say look at the so we have P open
11:31parentheses A/B closed parenthesis
11:34that's we read that as the probability
11:36of a given B conditional on B so the
11:40probability that a stock earns a 10%
11:43annual return given that has it has
11:45earned 9% in each of the two previous
11:48years then we can use that equation
11:50right there to get that
11:55probability all right so remember I said
11:57in the losos introduction we're going to
11:59have to do some rules so here's a
12:01multiplication rule and we're going to
12:02have to worry about whether or not we
12:05have a a joint probability which is the
12:09probability of two events happening
12:11together here's the example that we've
12:13come up with both both there's the key
12:15word in there both gas prices and bus
12:17Fair increase at the same time so the
12:21multiplication rule states that the
12:23probability of A and B so we would read
12:26that the joint probability of A and B is
12:28equal to the the probability of a given
12:31B which we just did on that previous
12:33slide times the probability of B so
12:36that's uh that's joint probability
12:39however what we need to be aware is if
12:41these two are independent events meaning
12:44that they are not conditional upon each
12:47other then uh the uh multiplication rule
12:51tells us that we just take the
12:52probability of a times the probability
12:55of
12:57B here's a good example that we've come
13:00up a bag contains 16 blue balls and 14
13:04yellow balls two balls are drawn from
13:06the bag at
13:08random one right after the other without
13:11replacement what will be the joint
13:13probability of both balls being
13:16blue now I was taught by a good math
13:18teacher Way Way Back hundreds of years
13:20ago to take a look at the question and
13:23see if you can figure out what the
13:24answer might be can you give me an
13:26estimate without knowing anything about
13:27probabilities I mean look we have 30
13:29balls right in this bag 16 of them and
13:3214 of them are two different colors so
13:34about half uh are uh each color in there
13:38so if you pick out one and then you pick
13:40out another half times a half so this
13:42has to be around 25% right so let's go
13:45ahead and do the actual math demonstrate
13:48the application all right so the
13:50probability that the first ball is going
13:52to be blue is of course 16 out of 30 but
13:55then since we're doing without
13:57replacement then we need to adjust this
14:00down to 15 out of 29 so all we do is
14:03take uh 15 out of 29 * 16 out of 30
14:07which is almost it's almost like the 50%
14:09time 50% that I suggested to you before
14:12I showed you this gray uh box so
14:1627 what was that 27.59%
14:22how about an addition rule we need to
14:25make sure we're aware of mutual
14:27exclusivity so if we have mutually
14:30exclusive events A and B and there's a
14:32good box up there in the top right A and
14:34B so all we're going to do is sum those
14:36two probabilities but for events that
14:39are independent or nonmutually exclusive
14:42so bring those two circles in together
14:45so there's overlap in there and what we
14:48don't want to do is we don't want to
14:49count the overlap twice so the addition
14:53rule for
14:55non-mutually exclusive events is exactly
14:58what it was for Mutual exclusivity above
15:02right probability of a plus the
15:04probability of B but then we need to
15:06subtract out that
15:09overlapping uh part of the two circles
15:11so we just subtract out the probability
15:14of A and
15:15B so here's a good uh question that
15:17we've come up uh CFA candidate
15:20probability that she gets the first
15:22question correct is 30% probability she
15:25gets the second question correct is 40%
15:29given the probability that she gets both
15:32questions correct is
15:3510% what is the probability that she
15:37gets the
15:39first the second or both questions right
15:43so we're going to say a plus b right 30%
15:47plus 40% that gives us that gives us 70%
15:50minus the 10% right so there's
15:5560% compare and contrast dependent and
15:59independent events so these are exactly
16:01what you might think dependent events
16:03are those that depend on each other and
16:05and independent events are those that do
16:07not so look at our dependent event
16:09example down there in the bottom right
16:10we just did this if you draw a blue ball
16:13without replacing it the probability of
16:15drawing another blue ball changes right
16:18that was the what were those numbers
16:20back there I might as well go back
16:23instead of the 16 out of 30 it becomes
16:2715 out of 29 that should make sense
16:29there independent events meaning that if
16:33you do something over here it's not
16:34going to have an impact of over here
16:36rolling the dice and flipping a coin and
16:40there's The annotation that we use there
16:42in the
16:44middle how about unconditional
16:47probabilities this is also known as
16:49marginal probabilities and marginal
16:51probabilities are going to become really
16:52important as we move through level two
16:56and maybe even a little bit in level
16:57three when we talk about correlations
16:59and causation and and we want to see
17:02exactly what that marginal probability
17:05of you know some event occurring like
17:08stock price going up
17:10Etc probability that an event occurs
17:13without taking into account any other
17:16preceding event ah so unconditional
17:19probabilities are not conditioned right
17:22and so we refer to those as Standalone
17:26events probability that a stock earns
17:2810% annual return without considering
17:32the preceding annual returns
17:34unconditional
17:37probability now here's another
17:39probability rule total probability rule
17:42explains an unconditional probability of
17:45an event in a series of mutually
17:48exclusive and exhaustive scenarios so
17:50here we're going
17:53to compute the probability of event a
17:57occurring given some other stuff over
18:00here so there's the notation but let me
18:02go ahead and show you exactly what we
18:03mean here so notice down the left-hand
18:06column we have two states of the economy
18:09no recession and a recession so one is
18:13the complement of the other there's that
18:15c in the superscript so probability of
18:18no recession 70% probability of a
18:23recession is 30% and then we can go over
18:26and annotate the stock price performance
18:29rise and fall so there's the slash given
18:32the complement or not and so there's the
18:34associated probability so what is the
18:37total probability of a stock rise so
18:42here we go all we're going to do is take
18:44I mean this is really just a weighted
18:45average so uh we're going to take
18:50the
18:5180% of the probability of an economic
18:55state right so 80% of of 70% and then
19:00we're going to add to that the 30% times
19:04the 30% of the probability of that stock
19:08performance if it goes up and then the
19:10probability of that economic state of
19:1230% so we get 65%
19:19there let's move on
19:22to what does all this stuff mean when we
19:26put together a portfolio because we're
19:29going to have if we have let's just say
19:30we have a three asset portfolio we're
19:33going to have asset one asset two and
19:35asset three each of those assets is
19:37going to have its own outcomes right
19:41based on states of nature so they're
19:43each going to have expected values and
19:45then they're each going to have their
19:47variability around that expected value
19:50so we want to calculate the mean and
19:52standard deviation so this losos asks us
19:55to hey apply a little bit of this
19:58probability concept that we've just
19:59learned to portfolio analysis so I'm
20:02going to go ahead and read this very
20:04first arrow point expected value of a
20:06random variable is the probability
20:08weighted average of the possible
20:11outcomes all right so think of
20:14probability of X time X1 or X1 * X1
20:18there is the expected return on a
20:22portfolio uh based on probabilities and
20:25each of those outcomes occurring for IND
20:28idual stocks then you can add it you can
20:30add it uh to portfolios of
20:34stocks and then the variability around
20:37that expected value there's the formula
20:40for variance and we just going to take
20:42the square root of the variance to get
20:43the standard deviation and what we're
20:45doing is we're taking each possible
20:47outcome there's the x minus the expected
20:50value of x which is the mean and then
20:52we're squaring it and we're squaring it
20:54uh as my students always tell me uh to
20:57get rid of the negative and that's
20:59that's true uh but we're we're uh we're
21:02we're squaring it to get rid of the
21:04negative so and there's a comma after
21:06that and the students don't ever give me
21:08the sentence after the comma the comma
21:10after it tells us that what we want to
21:12do is make sure that the minus numbers
21:15don't exactly offset or offset at all
21:18the plus number so we Square it to get
21:20rid of the
21:21negative and look at that last arrow
21:23point standard deviation is easier to
21:25interpret uh because it is in the same
21:28units as the random variable yes
21:30remember the variance is a squared term
21:33and uh your husband or wife or your most
21:37important person in your life doesn't
21:39ever tell you to stop at the food store
21:41on the way home and get 12 squar pounds
21:44of potatoes or some other kind of uh
21:48crazy kind of
21:50uh of a suggestion so we take the square
21:53root because it's makes sense right
21:55easier to interpret but it makes sense
22:00all right so let's go ahead and use this
22:01into uh the losos that asks us to
22:04calculate so we've got probabilities
22:06down the left- hand side remember what
22:08did we say earlier those probabilities
22:09must sum to one so 20 and 30 and 50 that
22:12gets us 100% right and so here are the
22:16sales so we have a 20% chance of having
22:1950 million we have a 30% chance of 40
22:22million and 50% chance so what is the
22:25standard deviation of sales and this is
22:27kind of a trick question question
22:28because before you can compute the
22:30standard deviation of sales you have to
22:32compute the expected value of those
22:35sales so all we're going to do is
22:37multiply column one by column two right
22:4120 * 50 30 * 40 50 * 30 and we'll sum
22:45those so that's a that's a probability
22:48weighted average and and we get 37
22:52million now what we need to do in order
22:55to get the variance of sales or the
22:57standard deviation of sales is that we
23:00need to go ahead and take each possible
23:04outcome so 50 is our first outcome minus
23:09the expected sales what was that 37 in
23:11that previous one yeah minus 37 we're
23:13going to square it and we need to
23:15multiply it by that probability of 20%
23:17and we do that all the way out to the
23:19end sum all those and we get uh we get
23:2361 million and we take the square root
23:25and we get around 8 million so we have
23:29you ready for this we have an expected
23:32value for sales of 37 million now the
23:36way we interpret that as good financial
23:38analyst is that of all the possible
23:41outcomes in millions of dollars of sales
23:44next year the most likely you ready for
23:47this the highest probability of
23:49occurring is that 37
23:52million now of course we want to ask
23:55oursel the question all right if we say
23:57we expect to get 37 7
23:59million let's follow up with well how
24:02wrong can we be can we be this wrong or
24:05can we be this wrong that's what
24:07standard deviation tells us it's a
24:09measure of risk so on average we're
24:12going to be about $8 million off on our
24:16estimate that's an that's what standard
24:18deviation tells us see the 61 million is
24:21a squared term so if you say 61 million
24:25is the variance you know I'm at least
24:27I'm scratching my head saying I'm not
24:28quite sure what that means take the
24:30square root now now we know that when we
24:33expect 37 million we're not going to be
24:36surprised if if the uh actual sales next
24:40year is you know let's say eight above
24:42or eight below now we'll go ahead and
24:45talk about confidence intervals at
24:47length and we'll need that standard
24:49deviation to do that but that's a
24:51conversation for another
24:54day all right how about conditional
24:57expectations all right right so the
24:59expected value of an investment given
25:01certain set of real world events so this
25:05goes back of course this goes back to my
25:07football example real world event is oh
25:10is it snowing or sleeing or raining or
25:13is the sun shining given a set of
25:16certain real world events so of course
25:21the expected value of an investment is
25:23impacted by lots and lots of these real
25:26world and events like
25:28what kind of new product lines is our
25:31competitor
25:33introducing what does the government say
25:35about our decision to take over another
25:38company do we have a monopoly is the
25:40government going to come in and uh offer
25:43up some kind of a lawsuit for antitrust
25:46violation or are institutional investors
25:50like financial institutions are they
25:52going to be required to increase their
25:56reserve requirements or decrease their
25:58reserve requirements from a government
26:01mandate or how are these financial
26:03institutions going to react all of those
26:06events right real world events are going
26:08to
26:10impact so what do we say there in that
26:13third arrow point total probability rule
26:15is very useful so there's the
26:18unconditional expected value is the sum
26:21of the conditional expected values so
26:25let's take a look at an example 20%
26:27chance that the government will impose a
26:29tariff on imported cars all right so
26:32company that assembles cars locally
26:34expects 14% return if the Tariff is
26:38imposed 11% if the Tariff is not imposed
26:41what is the unconditional expected
26:44return so once again this is just a
26:46probability weighted so we have a 20%
26:49chance of getting 14 which means there's
26:52the implication right an 80% chance of
26:54getting 11 you take a weighted average
26:56and you get uh somewhere around 11 and a
27:01half uh this is one of my favorite
27:03slides because we are going to use this
27:05again and again and again especially in
27:08level two especially when we have a
27:10discussion on derivative Securities and
27:13maybe option pricing and and maybe even
27:16swap valuation and maybe some other
27:18things out there but let's go ahead and
27:19look at a tree diagram so we start here
27:22time period zero right that's the over
27:25there on the left and then we branch and
27:27we say something like you know what of
27:29all of the possible outcomes let's
27:32suppose that we break it down into two
27:35so that we can say something like hey
27:37you know what something's going to
27:38happen up here and something else is
27:40going to happen down here the example
27:42that we've come up with in this slide is
27:45tossing a Fair coin twice Fair meaning
27:50that I guess it's a heads or a tails and
27:52it's not uh it's not weighted some way
27:54or so that the flipper can uh you know
27:57increase his or her probability of an
27:59outcome so a Fair coin toss those of you
28:03who are football fans I encourage you to
28:05go look up the Jerome Bettis uh coin
28:07flipping event in uh in the playoff game
28:10oh I don't know how many years ago that
28:11was 10 years ago it's it's super
28:14funny all right so note what we have we
28:17can flip the coin and we have a 50%
28:19chance there's the 0. five 50% chance
28:21there's the0 five of it being a heads or
28:23a tail so this is a tree right so there
28:25are the branches and then if we flip it
28:27again you know we're going to do the
28:29same thing and we have those equal 50%
28:33uh chances and so look on the middle
28:35columns out there so what is the chance
28:37that we have two heads what's the chance
28:40that we have two tails what's the chance
28:42that we have heads then Tails or tails
28:45then heads so these are all of the
28:47possible outcomes and this is a visual
28:50representation of flipping a coin and of
28:52course those probabilities all have to
28:54be 25% but you knew that at least before
28:58maybe before even watching this slide
29:00deck but then that example that I gave
29:02you 50% earlier on all right so look at
29:05the very bottom so first toss second
29:07toss there are the outcomes possible
29:09outcomes and then the
29:14probabilities all right let's do a real
29:16live uh example suppose the prospects
29:20for recovering a principle on a
29:22defaulted bond issue depend on which of
29:24two economic scenarios Prevail so
29:26scenario one probab ility of 60% which
29:30means scenario 2 has a probability of
29:3240% so that scenario one will recover uh
29:37in will result in a recovery of 80 cents
29:42now that's per $1 right with a
29:43probability of
29:4535 70 cents with a probability of 65 all
29:49right so scenario one we've got a couple
29:51of probabilities layered in there
29:53scenario 2 has a probability of 40% will
29:56result in a recovery of only 60 cents
29:59with a probability of 70 and then 30
30:02cents with the probability of 30 so what
30:05is the expected recovery closest to so
30:08here's our here's our uh tree our
30:13decision tree so we have 60% going up
30:1640% going
30:19down and then we hit scenario one and
30:21scenario two then we have different
30:23probabilities right we have 35% and 65%
30:26then 70 and 30 so there are the recovery
30:30amounts there's the 80 and the 70 the 60
30:33and the 30 and so all we're going to do
30:37is we're going to estimate the recovery
30:41today expected recovery today so we're
30:44going to take a weighted average
30:46probability of that 80 and that 70 so 35
30:49and 60 that gets us
30:51735 and then scenario two there's 70 and
30:5430 and uh we get 51 so that was based on
30:59those probabilities out there on the
31:01right hand side of the tree now we have
31:03to you know I'm going to say discount
31:05but that's really not the right term
31:06we're going to assign those
31:08probabilities of 60 and 40 before we
31:10know which scenario actually occurs so
31:13we get
31:15645 let's go ahead and continue our
31:17discussion on portfolio returns so look
31:20at this losos calculate and interpret
31:22expected value which is a mean variance
31:24standard deviation co-variance and
31:27correlation so we have lots and lots of
31:31um random variables that are going to go
31:33inside of this Los so let's go ahead and
31:36start with a simple
31:37one uh calculate the expected return on
31:41a
31:41portfolio now of course the expected
31:44return on a portfolio depends on what is
31:47the expected return on each of the
31:50individual Securities inside of that
31:52portfolio you know for example if we
31:54have if we have three stocks in our
31:57portfolio and the expected returns are
31:5920 25 and 30 no matter how we wait them
32:03the expected return on that portfolio is
32:05going to be between 20 and 30% but if we
32:08have three treasury Securities and the
32:11expected return on each of those is you
32:13know one one and a half and two no
32:15matter how we waight them we're going to
32:17get between well what did I say one one
32:19and two% so the W in the formula is the
32:23proportion of dollars or currency
32:27allocated to that individual stock so
32:29let's go ahead and read that Circle
32:31point it's a weighted average of the
32:33expected Returns on the component
32:36Securities so if we have three assets
32:38let's say asset one two and three they
32:40have expected returns of 10 20 and 30%
32:44well the expected return on the
32:45portfolio is going to depend on how we
32:48allocate all right so this is an asset
32:50allocation question what is our weight
32:53one our weight two and our weight three
32:54now of course those weights have to sum
32:56to one
32:59and so what did I say you know if we
33:00have 10% in weight in in asset one and
33:0310% in asset two that means we're going
33:05to have 80% in asset three which was the
33:08highest of the three so we're going to
33:09have a higher expected return on the
33:12portfolio so before I go on remember
33:15that the expected return on the
33:16portfolio depends on you know which kind
33:18of assets you have and then how you
33:21allocate your wealth across those
33:25assets now the variance of the return on
33:28the portfolio is really kind of a cool
33:31function and back in the old days when I
33:34would teach this to my MBA Investments
33:36class I would take my students through
33:37all the steps and I would come up with
33:39this equation there that's in that's in
33:41the gray and I would say you know like
33:44here it is and the students would say
33:45you know what Jim do we really need to
33:47know all that stuff my answer of course
33:50was yes of course the more you can cram
33:52into your brain the better off you are
33:54but let me just go ahead and tell you
33:56what I my MBA students now I say hey
34:00look remember back when we did the
34:03variance of the returns for an
34:04individual security we took each
34:07expected outcome minus the mean outcome
34:10and we squared it remember we squared to
34:12get rid of the negative well if you put
34:15this equation there inside of that uh
34:19expected return formula and the standard
34:22deviation formula you get lots and lots
34:24of subtractions and then lots and lots
34:26of squares and when you cancel and
34:29collect terms and you do a bunch of
34:30algebra this is what you get all right
34:33so the variance of a two asset portfolio
34:38is not the weighted average of the
34:42standard deviations or the variances of
34:44the component Securities and so that's
34:46the important first lesson here so what
34:48we're going to do is we're going to take
34:51the weight in the first asset or asset a
34:53in this case and we're going to square
34:54it multiplied by the variance and then
34:57we're going to take the weight in the
34:59second asset and we're going to square
35:00it multiplied by the variance and so
35:03that takes care of all right I want you
35:04to look on the right hand side of the
35:06equal sign and then to the left of the
35:08plus
35:09sign that takes care of the weighted
35:12individual riskiness of asset
35:15a the next thing to the right of the
35:19first plus sign and to the left of the
35:21second plus sign that takes care of the
35:24weighted riskiness of asset B
35:28so you have the variance of a portfolio
35:30depends on the riskiness of one of a and
35:33the riskiness of B and then there's a
35:36plus sign over there and I always tell
35:38my students to think of this as the co-
35:41riskiness because A&B do not operate
35:45inside of a vacuum right what do we know
35:48when you go to the floor of the New York
35:49Stock Exchange that stock returns and
35:52stock prices tend to move in the same
35:55direction so if everyone's happy and all
35:58stock prices are going up then one
36:00individual stock is probably going up as
36:02well so we need a measure of that co-
36:06riskiness and that's what we've written
36:08up there in that Circle point right so
36:10the variance of a portfolio's return is
36:12a function of the variance of the
36:14component assets I just did that for you
36:16as well as the co-variance between each
36:19of them so over on the right hand side
36:22we have plus two because a co-varies
36:26with b and b covaries with a not not
36:30because there's two Assets in the
36:31portfolio so the two is because A and B
36:35and B and a covariate against each other
36:38by the same amount so you multiply it by
36:40two then we multiply by the weight in a
36:43then the weight in B and then we
36:46multiply by that co-variance term out
36:48there so let me goad and give you some
36:51introductions to co-variance covariance
36:53is a measure of the strength and the
36:56direction of a
37:00relationship think about it this way if
37:02we have stock a and stock a does this
37:05here it is in time period zero and
37:07here's Stock B in time period zero if
37:09stock a moves like this Stock B could do
37:13a bunch of things it could go backwards
37:15it could go forwards a little bit it
37:16could go farther it could go over here
37:18to the side I don't know if it could
37:19jump over here that way but it's going
37:22to measure covariance is going to
37:23measure how B reacts to What a
37:29does so we get the variance using that
37:32formula and then we just take the square
37:34root of that variance to get the
37:36standard deviation of the portfolio so
37:39let's go ahead and take a look at an
37:41example right here we have two mutual
37:43funds A and B 60% invested in a 40%
37:47invested in B and the covariance is
37:50given as
37:551.44% so further information we've got
37:58the expected return on A and B is 15 and
38:0110% and the standard deviation is 20 and
38:0515% so what we're going to do is say ask
38:08the question what is the expected return
38:11so the expected return has to be between
38:1315 and 10 right standard deviation of
38:16returns you're tempted to say you're
38:19tempted to say that it has to be between
38:2120 and 15% however I want to caution you
38:26it's probably never going to greater
38:28than 20% I can live with that but it
38:30might be less than 15% depending on the
38:33degree of co-variance up
38:36there well let's see what happens in
38:38this example so there's our weighted
38:40average so all we do is take 60% of 15
38:44and 40% of 10 that's a simple weighted
38:46average expected return so what did I
38:48say had to be between third uh between
38:5010 and 15 there it is
38:54133% ah but for the portfolio variance
38:57let's make make sure we go through this
38:58correctly and please don't make the
39:00common mistake that some of my
39:02undergraduate students do my MBA
39:04students don't do this too often and I'm
39:06hoping that you won't do this don't
39:08forget to square those first four
39:09numbers right 60% squared time 20% squar
39:12plus 40% squar times 15% squared the
39:16last thing that you want to do on the
39:17exam you know in the stress you know
39:19you're under time stress and you go
39:20through and you forget to square one of
39:22those so take a deep breath and uh and
39:25take your time make sure that you do
39:28this and that's my advice all the time
39:30in fact next time you watch uh uh an NBA
39:33basketball game watch watch one of these
39:36dudes or the WNBA watch one of these
39:38women they get the foul shot and what
39:39did they do they take a dribble or two
39:41and they take a deep breath this is what
39:42you got to do in these kinds of problems
39:43you know it's long mathematical stuff so
39:45take a deep breath and make sure you go
39:48through your routine and you follow
39:49through so you can make your foul
39:51shot so we get what about 2 and a half%
39:54for variance take the square root of
39:56that and and uh you get
40:0515.78% it might be between 15 and 20%
40:10and that's the case in this example but
40:13it might even be less than 15% because
40:16of the beauty of diversification now
40:18remember what I said I went like this in
40:20this case the co covariance was what was
40:23it 1.44% so when a went like this B B
40:27moved in the same direction but suppose
40:30that when a goes like this B moves in
40:32the opposite direction so that the
40:34covariance is negative just for uh fun
40:38and Giggles put a minus sign in front of
40:41the
40:420.0144 then you bring the minus sign out
40:45and that turns the plus sign into a
40:47minus sign and you're going to get a
40:48much lower variance take the square root
40:50of that and you're going to have a
40:52standard deviation of oh let me make up
40:54a number you know 12% or 9% or 11%
40:58that's the beauty of diversification
41:00we'll talk about this at length as we go
41:02through levels one and two and three but
41:04diversification depends on the
41:08co-variance between those two
41:12assets all right here's the uh here's
41:14the formula so calculate and interpret
41:17co-variance and correlation coefficient
41:19so uh what I want you to do is look uh
41:22look at the formula for variance that's
41:24I copy and pasted that from an earlier
41:26slide so if we do the variance all we're
41:28doing is taking uh each possible outcome
41:31y minus its mean and then we're squaring
41:34it right I just left that term over
41:35there so y minus the expected value of y
41:39times y minus the expected value of y
41:42That's squared right well the
41:44co-variance between X and Y is very very
41:46similar but instead of the variance
41:49where we're squaring it and essentially
41:51we're saying in the variance formula
41:53we're saying hey how much does y vary
41:56against itself we call that the variance
41:58but with co-variance now we're saying
42:00how much does X vary against y so
42:04there's the equation for the co-variance
42:06we take each possible outcome for x
42:09minus its mean value and we multiply
42:12that by each possible outcome for y
42:15minus its expected
42:17value so look at the example we have in
42:20that second Circle point we might be
42:22interested in the degree of co-movement
42:24between interest rates and inflation oh
42:26my God that makes perfect
42:29sense now of course what we can do is
42:32that we can throw probabilities in there
42:34and you shouldn't be surprised that
42:36we're going to throw probabilities in
42:38there and this is just going to be a
42:40probability
42:42weighted uh covariance between the
42:45assets so this equation here and that
42:48equation right there they're the same
42:49equations but instead of x's and y's we
42:53use the notation here look at the second
42:55Circle point if we have two assets I and
42:57with returns i r sub i and r subj so
43:01we're taking R sub I minus the mean of R
43:05and then R subj minus the mean of J and
43:09then notice that right there in the
43:10middle we have the probability of each
43:14of those possible outcomes right so let
43:16me read those B uh bolded points up
43:19there based on the probability weighted
43:22average of the cross products of the
43:24random variable deviations okay that
43:26makes sense so let's go ahead and take a
43:29look at an example we have economic
43:31growth so two two choices here less than
43:33four greater than four we have 40% and
43:3660% chance so return on stock a and
43:41return on Stock B there's the table
43:43we've seen a couple of tables like that
43:45before what is the co-variance of
43:47returns so here is an explicit example
43:51of how we're going to do this but first
43:54we need to do the expected return on
43:56each stock so stock a we've done this
43:59before right we get 572 Stock B we get
44:034.40% so then going down that second
44:06column we take each outcome 2.3 and
44:10eight subtract that from 5.72 there's
44:13the mean that we just
44:15calculated and uh in column three we
44:18take each possible outcome six and a
44:21half and three subtract that from four
44:254.4% we get the - 7 and -3 multiply that
44:30by the probability so make sure you go
44:31back here see those probability the p uh
44:36of R sub I is and close parentheses that
44:41doesn't come in until the
44:43end and then you get probability
44:45weighted product so the co-variance is
44:524.79% calculate so we just calculate how
44:54about interpret it ah let's interpret so
44:57the the minus sign the minus sign tells
44:59us that when a goes this way B goes in
45:01the other direction now not all the time
45:04now that doesn't mean that every single
45:06day that there's going to be that
45:07inverse relationship but on average
45:10right and we're waiting the averages by
45:12that probability on average that sign is
45:15negative which means they tend to move
45:18in opposite
45:22directions ah this is what I just said
45:24right positive co-variance Returns on
45:26both assets tend to be on the same side
45:29right negative and zero Co zero
45:31covariance means that when when this
45:33thing goes like that then this thing
45:36could do anything it could go back it
45:37could go forward it could go sideways it
45:39could jump up it could do anything I
45:40mean it's not going to do all those
45:41other things right but uh uh the idea of
45:45a zero covariance means that the two
45:47assets are independent and the the
45:49reading uses the word unrelated which
45:51means that when one does something you
45:53really can't predict what the other one
45:55will do
45:58now remember what I said just a few
45:59moments ago that the variance is a
46:02squared term and you never buy 12 square
46:06pounds of potatoes at the food
46:09store
46:11well I did say to you that that's a
46:14difficult number to interpret well
46:17covariance is similar in that because
46:20even though you're not squaring the
46:21number right you're multiplying it by
46:23another number so it's it's kind of
46:25squared it's not really squared but but
46:27it's kind of squared and so here's
46:29here's what happens with co-variance
46:32when you compute the co-variance between
46:34any two random variables you can come up
46:36with almost any number like you know a
46:39plus 16 million or or a minus 400,000 or
46:440.52 and so just like we had written in
46:47an earlier slide that variance is
46:51difficult to interpret so so is
46:53covariance so statisticians are super
46:56smart they say okay if we can't uh
46:59really understand what the absolute
47:01value of a covariance term means why
47:04don't we standardize it so think about
47:07what I said I've done this before on
47:09slides is that you know the covariance
47:11can be it can be any number can you see
47:13my hands way way out here but what we're
47:15going to do with the correlation
47:16coefficient is if we take the
47:18co-variance and we put it in the
47:19numerator and we divide by the product
47:22of the standard deviation of each of the
47:24two individual Returns what doing is
47:27we're putting covariance into a vice and
47:29we're squeezing it and we're squeezing
47:31it all the way down to
47:33some pable range right co-variance can
47:37be almost any range correlation
47:39coefficient can be minus one or plus one
47:43or anything in between so think of the
47:45correlation coefficient think of it as
47:47the ratio right that's the definition of
47:50covariance divided by the product of the
47:53two standard deviations but I want you
47:54to think about it as a standard IED
47:57covariance and that means that it's
47:59going to be between a minus one and a
48:03plus one so this gives us a better sense
48:05of when one goes like this and one goes
48:07like that exactly what that
48:10means a strong positive linear
48:13relationship indicates a perfect linear
48:15relationship which would be one so if if
48:18we have a correlation coefficient of .9
48:21that means when a goes like this that on
48:24average B is going to go 90% of the way
48:27now don't ever tell a stats Professor
48:29that I said that but that's that's a
48:31vague interpretation over time in other
48:34words if the correlation coefficient is
48:35a minus8 and a goes like this that means
48:38B is going to go 80% in the opposite
48:42direction so now we can wrap our arms
48:44around correlation and say magnitude and
48:48direction of a relationship and of
48:51course zero correlation I mean look if
48:53covariance is zero in the numerator then
48:55correlation has to be zero right no
48:57linear
48:59relationship all right so let's go ahead
49:01and take an example interest rates and
49:03inflation uh historical date of five
49:06years co-variance is a minus uh standard
49:09deviation is 5 a half and 12 so
49:12calculate and interpret so all we're
49:14going have to do is standardize
49:15standardize that covariance term by
49:18multiplying by the product of the two
49:21standard deviations so since the
49:24covariance is minus is has a negative
49:27sign in front of it then correlation
49:29coefficient is going to have that same
49:32minus so what do we say first is that
49:34hey when uh interest rates go up
49:38inflation goes down or vice versa
49:41remember correlation doesn't tell us
49:42anything about
49:43causation and now it's 0.11 so that's
49:47relatively close to zero so we're going
49:49to say that that relationship while
49:51negative is not super strong so what do
49:54we conclude there at the bottom the
49:57interest rate and inflation are
49:58negatively correlated with each other
50:00but it's not super strong if that turned
50:02out to be a minus .91 or 081 then we
50:05would say oh my gosh these two things
50:07are super
50:10related now the reading shows What's
50:13called a co-variance matrix uh when I
50:17was in graduate school my professors
50:19called it the variance co-variance
50:21Matrix and I'll explain that to you here
50:22in just a second and notice that we have
50:24colorcoded these things so we have have
50:27asset a b and c going down the First
50:30Column and asset a b and c going across
50:33the first row so I want you to look at
50:35The Two Reds so the co-variance between
50:38b and a has to be the same as the
50:41covariance between A and B so The Two
50:44Reds are the same so the covariance
50:47Matrix is really a mirror image of each
50:49other same thing with the two greens
50:51right the co-variance between C and A
50:54has to be equal to the co-variance
50:56between A and C and then of course the
50:58purples uh between uh B and C as well
51:02now look at the blues going down the
51:04diagonals the co-variance between R A
51:07and R A well we could rewrite that
51:09that's the variance of a and then the
51:11variance of B and then the variance of
51:16C so notice the second arrow point a two
51:20asset portfolio would have a similar 2x
51:22two Matrix all
51:25right what you can o do look at that
51:27last Circle point a correlation Matrix
51:29can be created to represent correlations
51:32between and among the various
51:36assets all right let's get back to uh
51:39let's get back to conditional
51:41probability so what was that example
51:42that I gave you that uh we were trying
51:46to determine the outcome of a football
51:48game this weekend and the two teams were
51:50evenly matched right but what we found
51:54out is that even though they're evenly
51:56matched one does better in the snow and
51:58SLE and rain and one does better in the
52:01sunshine so we call that a conditional
52:04probability that hey wouldn't it be
52:06awesome if we knew what the weather was
52:08going to be like the instant before the
52:10game started then then we could uh we
52:14could update our probabilities well
52:17that's what bay does and this goes back
52:19to oh I want to say 1770s or 1780s so
52:24the reason that I mentioned that to you
52:26is because fascinating that you know
52:27stuff that happened hundreds and
52:29hundreds of years ago the CFA Institute
52:32think is still important today so look
52:35at that first point there base formula
52:37is used to calculate an updated
52:40posterior probability given a set of
52:43Prior probabilities all right so the
52:45Reverend tbay 1775 or whatever year
52:49widely used basian methods of
52:51statistical
52:54influence so notice that third point
52:56there it allows us to turn around
52:59conditional probabilities and I like to
53:01think of it as this is taking new
53:05information and adjusting the
53:07probabilities you know like if it's
53:09Wednesday and the football game is going
53:11to be played on Sunday well we can
53:14compute a conditional probability of an
53:17outcome based on you know the relative
53:19strengths of two teams and the Wednesday
53:23probability of a snow event or rain
53:26event or a sun event uh that coming
53:29Sunday but still we're estimating that
53:31probability with some degree of
53:33uncertainty but now as we get closer and
53:36closer to the game time we can update
53:39those
53:40probabilities and that's what this base
53:42formula does and there it is shown in
53:45the gray box down
53:47there so let's go ahead and take a look
53:49at these probabilities here or the
53:50notation uh PE these are prior
53:53probabilities the event a is some event
53:56known to have occurred and the
53:59probability of e given a is the
54:01posterior probability so let's go ahead
54:03and take a look at an
54:05example let's suppose we want to get to
54:07work on time so we have trains uh number
54:10of train Journeys 50% New York 30% Vegas
54:1420% DC so we have crazy work schedule
54:18right probability of a train arriving
54:20late in New York Vegas and Washington
54:22are 40% 3 35 and 25% so the first thing
54:27you need to do is shake your head and
54:28say oh my gosh why aren't these
54:30probabilities of the train arriving late
54:31like 1% and
54:342% all right so here's the question if
54:36the engineer picks a train at random
54:38what is the probability that it
54:40terminated late in New
54:43York all right so let's work our way
54:45through some of this notation so n is
54:48the event a train chosen at random
54:50terminated in New York V is the event a
54:54train chosen at random termin in Vegas
54:57and the same thing for w all right so
55:00we'll let uh we'll let L be the event a
55:03randomly chosen train AI arrives late
55:07but remember look at the top we're
55:09looking for the probability that New
55:11York is going to be late so all we do is
55:14go back here and extract these
55:17probabilities so in the numerator we're
55:19going to have uh 50% and 40% and then in
55:22the denominator in this base formula
55:24we're going to sum a couple of those
55:27probabilities the of course the
55:30numerator 50% time 40% and then the
55:33other two trains 30 * 35 and 20 time 25
55:36so we get a little more than half right
55:4156% now how about these counting
55:44problems so these counting problems
55:46involve determining the exact number of
55:48ways two or more operations can be
55:51performed together and so look at the
55:53example that we've chosen so we can
55:56choose seven stocks three small cap four
56:00large cap from a group of 50 stocks so
56:03think about this you got 50 stocks you
56:04know in this big bag and we're going to
56:06reach in and we want to choose seven of
56:08them right how can we determine the
56:12number of counting so there's some good
56:14issues
56:16here so let's go ahead and start with
56:20this n factoral which I think I said in
56:23the beginning of the slide deck uh I
56:25learned in fifth grade or seventh grade
56:27whatever whatever grade that was but all
56:29we're going to do n factorial it just
56:32simply means that we're going to start
56:33with that number and multiply by
56:35multiply Itself by every number that
56:37comes before it um on the timeline so
56:41look at the example there 3 factorial is
56:44simply 3 * 2 * 1 which of course is 6
56:50and there's the location on the ba2
56:53plus uh for the factorial let me go
56:56ahead and grab uh let me go ahead and
56:58grab my favorite calculator look under
57:01the three button there's a blue n you
57:04guys see that so if you want to go ahead
57:06and those of you have the 12C you can do
57:09that uh in
57:12factoral uh labeling is interesting in
57:15that it's going to allow us to assign
57:19group labels uh K times to a total
57:23number of items so look we have N1 and
57:26two and three all the way out to n k uh
57:30so what's going to happen we want to
57:32categorize n items into K groups where
57:37we have some predetermined information
57:39and so all we're going to do there's the
57:41equation there n factorial over
57:45um uh over all of those NS multiplied so
57:49let's go ahead and do an example here 10
57:51Stock suppose your wish is to assign
57:54three different labels such that we have
57:57highrisk stocks uh medium risk stocks
58:01and lowrisk stocks and you know on the
58:04exam you probably uh at least I would do
58:07this you know it's probably likely that
58:10you might read a sentence that says
58:11something like you know high risk medium
58:13risk and low risk is determined by hey
58:15what did we just do standard deviation
58:17of those stocks or maybe we want to use
58:21uh beta as we move along all right so
58:24there we go we're going to take the 10
58:26factorial right there's our 10 stocks
58:29and then we're going to multiply I'm
58:30sorry we're going to divide by well we
58:34have five high-risk stocks so there's
58:35the five factorial and then two uh three
58:39medium stocks there's the three
58:40factorial and then the two lowrisk stock
58:43so there's the two factorial so if you
58:45do all that I mean in the numerator you
58:47do 10 factorial you get some gigantic
58:48number right divided by a little bit
58:50less of a number you get
58:522520 ways
58:57now you can combine all different sorts
59:00of Securities um and let's go ahead with
59:04the added assumption that the order does
59:06not
59:07matter uh so what we're going to do once
59:10again is take the ratio of factorials
59:13but we need to subtract something in the
59:16denominator so let's take a look at a
59:17quick example here how can we choose
59:20three stocks from a portfolio consisting
59:22of 10 stocks how many ways can we choose
59:25so there is the there are the steps in
59:28the ba2 plus so all we're going to do is
59:30take the 10 factorial right and divide
59:32that by the 10 factorial minus the 3
59:36factorial times so you get 120
59:40there ah permutations ah let's suppose
59:44the number of possible ways where the
59:46order is super important so all we're
59:49going to do then is take the factorial
59:52divided by n minus the r which is the r
59:56items so let's do three stocks from a
59:59portfolio consisting of 10 stocks in
1:00:02order to execute a sale but the order of
1:00:04the sale is Paramount so you get 720 so
1:00:07what do we get on this one back here 120
1:00:10but now we get
1:00:13720 so here's the good question here I
1:00:15would take a picture of this and put
1:00:17this in uh in my camera I'm not my
1:00:21camera in my phone under uh level
1:00:24one slides uh to
1:00:28memorize so this is a good one to help
1:00:30you uh read the question stem so as
1:00:33you're reading the question stem if
1:00:34you're asked to do some of these things
1:00:37then this uh slide deck will tell you
1:00:40what to
1:00:41use and that takes us through all these
1:00:44boy I'd love to pick out one of these
1:00:46that's super important but I will say
1:00:49that I really like just exactly what I
1:00:51said you know number you know calculate
1:00:54and interpret conditional probability so
1:00:56in order to do that you need to
1:00:58understand the first three and then of
1:01:01course look at that one in the middle
1:01:02expected value variance standard
1:01:04deviation covariance and correlations of
1:01:06portfolio returns that's important
1:01:08moving forward
1:01:13[Music]

About this transcript

This page contains the full transcript of Probability Concepts (2025 CFA® Level I Exam – Quantitative Methods – Module 3) by AnalystPrep, generated from the public captions YouTube serves with the video. The transcript has 9,919 words across 1,406 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.