Statistics 101: Normal Distribution and Stock Risk
ฝัง
- เผยแพร่เมื่อ 1 เม.ย. 2017
- In this video we use our knowledge of the normal distribution to compare the risk (variance) associated with two sets of familiar stock returns.
My playlist table of contents, Video Companion Guide PDF documents, and file downloads can be found on my website: www.bcfoltz.com
One of the best analysis I've ever watched. Very good job!
Oh, what a calm and comfortable voice to listen to. I want more of this!
Brandon glad to see you back. Was so happy to see a notification from your channel about new uploads. I am spreading the word about your videos and more people will be subscribing. Keep up the great work!
It's great to see you back. The best stat teacher on TH-cam
Hello. I've been using the NORM.DIST function incorrectly for weeks! You saved me man! You're a legend! Thank you forever!
I got an A in my final exam on Stats! Thank you for yr help and encouragement.
Glad to see you back. Great video. Thanks a lot!
Thank you for adding the stock comparison case study. It really tied together the theory with real-world application/insights and helped to embed that in my mind.
One of the best till now
Appreciate the video, watched this for class and the excel functions were clearly shown on how to do this. The distribution explanation, the way you explained it drove the point home as to why this is helpful for stock analysis, or analyzing your portfolio. Thanks !
I love this exploration so much. I love math and stats because of you.
Brandon, thank you so much for this video. I am an MBA student and needed a refresher. Your video was the perfect foundation for me. I really appreciate you making this available. It should be mandatory viewing for all data analytics students!
Excellent analysis! You make complex concepts so simple! Thanks :-)
Excellent Video. With a practical example, it's easy to understand how risks is calculated. Thank you.
Thank you for the positive words at the end!
am so glad you're back. I was genuinely worried you wouldn't
Thnk you so much Brandon - the practical examples you provide are so helpful and clear!
You are welcome and thank you!
Really nice and easy for clearing the concept. Thanks
Great Lecture. Great Analysis. Thank you..
Hi Brandon. Thanks a lot, I am working on coming up with intraday trading strategies, this video will help me keep the volatile stocks at bay.
Great job making this video! I had an unconventional way of passing statistics...forex trading..I started trading at Agea and just then I realized what statistics and probability was all about. I just wanted to share this with your viewers,coz i think it might be helpfull:))
Great video explaining Normal distribution in a very pedagogic way. Will use my new knowledge when instructing classes.
Thank you. Your video was very helpful.
Hi Brandon, Just happened to watch this video and its really nice. Well explained. Thank you so much.
Hello sir!
I am student of Master in Psychology. I am struggling with stats since past 3 years. And have failed in stats 5 times till now. I am watching your videos to prepare myself for sixth attempt and this video helped me to understand concept which I was not able to understand. Even if I fail on next attempt I will always be thankful to you. Greetings From India.
HOW IS THAT EVEN POSSIBLE? After a second fail, the university would just not allow you to continue your carreer?
@@benderbendingrofriguez3300 here system is different, if we fail in a subject it's we call it k.t. we are allowed to have maximum 1 k.t. in semester.but we have to clear it before final semester or our degree get cancelled.I had K.T. in first sem. And I have cleared it also I am working since last 2 years, thank you again!
You are the best and i give you again 5 star
Thank you very much......was searching for this and u provided....👍👍🙏🙏
Awesome video!
Excellent...Thank you so much..!
Thank you for watching! 🍎
Glad to see you in action :-)
Thanks! Sometimes real life gets in the way. But I love helping my viewers as much as possible.
Wow you're back!!!! I thumbed it up b4 I even checked out the video!!
Thanks! Working on some new stuff. I really appreciate the support from you and everyone else. Keep on learning!
Thank you!
Thanks ..its a great info :)
Last thing you gave us was logistical regression, however you never seemed to complete it. Looking forward to any new vids we might see😁
awesome awesome sir
Excellent!!!
Thank you so much!
thanks so much!
.
Thank you
.
Great video, learnt a lot thanks!
Any idea where I can get data to create my own distribution?
Hello Brandon! Thank you very much for this video. I love the way you explain statistical concepts on your channel.
I wonder if you could help me with my dilemma. I've been trying to model stock prices using normaly distributed daily fluctuactions. Using Python, it's not a big deal. However, I've noticed some people using lognormal distribution of returns when modelling similar stock price series.
My question is: is this latter approach more suitable than mine or is it just an alternative way to model the data?
Thank you very much for your effort. You are a true teacher.
3 years later, I would like to add that log normal returns better represent stock returns.
Thanks, I'm looking for the steps to draw the weibull graph on SPSS, help me
For Q2, shouldn't z=-1.15 be on the right side of -1.73 on the x-axis? The slide points it to the left side of -1.73. Could you please confirm?
Well explained and excellent presentation. I would like to know if the there's an excel template available for this presentation. I saw through your website and I didn't find. Thank you for teaching
Hi mate, the videos help a lot. Thank you sincerely. One question, why do you have to type in 1 stdev zscore in the normal.dis formula if the value u are looking for is zscore stdev 2/-2? Best C
i missed your videos man
Me too! Real life is a pain! :) More to come!
Thanks a lot for you video. Suppose that you need that a given stock reach in total 10% of return in the next 3 trading days; what would be the formula of the probability of reaching your target? thanks!
Can you tell me what objective metric would you need to justify a set of data having a normal distribution?
There are several tools available such as normal probability plots, etc. Software will also have Kolmogorov-Smirnov, Shapiro-Wilk, and Pearson's chi-squared test among others.
Dear Brandon,
Thank you for your helpful videos. I have a question about the practice questions you provided. I was able to get the correct answer by using percentages, but I was wondering if I needed to use Z scores instead?
Hi Brandon, I have a question concerning the probability distribution of Apple's daily stock returns, When feeding the probability density function of the Gaussian distribution with the mean, standard deviation and daily stock returns I get an asymmetric bail shaped curve instead of a symmetric one (the one in the video), how is that ? which value of x did you feed the probability density function ?
Hi
The stock data not mandatorily states that it is of symmetric nature. It is basically considered by Brandon to be Normally distributed data for basically explaining that how standard deviation in a normally distributed/any distribution can be considered as the average degrees of deviation from the mean, and associated it with the risk, that how in the financial world, the risk is all about the deviation from the most likelihood value and how to calculate the probability of that risk, given the data is normally distributed in this video.
How can I use Statistics for day trading to keep track of stocks to find out when to short or go long before a breakout
How do you include price split?....
While calculating the cumulative probabilities, why are we passing the mean as 0 and standard deviation as 1 instead of 0.0011 (or 0.11) and 0.0184 (or 1.84) respectively?
Never mind, got it. It's because we're converting the absolute percentages into Z-Scores.
But why did we graph the given data as a normal distribution?
Hello
Thank you for the wonderful video. in Q1, in Apple corporation, why do we take mean as 0, and 1 as std dev? I thought, we would take 0.11 as mean and 1.84 as std deviation. Can you help me understand this?
sir can u please post video on Time-series analysis
i have off topic question that what will happen if we do not follow the assumption of ANOVA ? (ANOVA Test)
Normal distributions are not the appropriate model to measure risk for stocks see "When Genius Failed" and "Black Swan" by Taleb
Agreed. I've read both of those books. As I mention in the video this is not about Finance per se. Just a tool to help stats students. However I think the end conclusion, that by analyzing returns one can tell which returns are more volatile, is valid. Certain equities swing more from day to day.
Brandon Foltz thanks for the quick response. Your explanation videos are always clear and i agree with your conclusion.
Thaaaaaaaaaaaanks a loooooooooooooot Sir
Thank you Mr. Brandon Foltz, for this wonderful, easy to grasp & apply series of your presentation. I invest on my own, (am a Quant). I have a question. I have 20 Years of quarterly Stock prices of Indian Companies. I found their (quarterly returns) Mean & Standard Variation, I RANKED the companies on the following 1)HIGH Probability of Gain {>5%}, 2)LOW Probability of Loss (
Well, despite the math do spit out some information regarding the performance and volatility of companies, in the stock market, fundamental (reading business 10k/10Q) is really crucial as in every stock there is a business running behind. Therefore the numbers alone can't necessarily tell much in my opinion. Always remember that the number is self is part of the business valuation there are still many factors to be considered such as management team, business view, etc.
No, you won't be able to beat the market using this approach. It would only work (over the next 20 years) IF the returns distributions over the next 20 years are exactly the same for all stocks as the last 20 years. The chances of this are very, very small.
The other problem is that stock returns don't follow a normal distribution curve so any theoretical risk and return probabilities calculated from a normal distribution curve will be inaccurate.
You are probably better off handing your money over to a good fund manager (or several good fund managers) than expecting to beat the market using flawed statistics like this.
At 20:53, why are we using 0 and 1? Shouldn't it be the mean and standard deviation, which are 0.11 and 1.84?
Wouldn't it be right to use Z- table to find the probability having found the Z_Value?
Could you please teach us how to make normal curve in excel using any data.. apple or GE etc. ??
I would like to learn that as well, please
Isn't this one of the situations in which it is a bit deceiving calculating the mean in an arithmetic way? For example, if the daily returns for a stock in a 30 day period were alternating like +10%, - 10%, +10%, - 10%,..., then the arithmetic mean return would be 0% but every 2 days we would be losing 1%. So, some of the premises used in this video to compare the stocks are a bit flawed, or have to be stated more precisely, right?
Just commenting since it reminded me of your video on the geometric mean.
Hello! If looking at returns over a period of time, such as a time series, then that might be the case. But a distribution of returns is a common measure of risk, or volatility, in any given security. Essentially looking at how "swingy" the returns are as a whole.
If we are looking for longer term analysis, we could use CAGR (due to its simplicity) to look at the compounded growth rate over a given period of time based on initial and current market value since it smooths the returns and ignores short-term movements.
But how do I know mu and sigma of population having only samples??
So what is the hypothesis test? isn't the returns referring to "average daily return" rather than simply daily return on a given day?
APPL daily returns are NOT normally distributed!!
A - MAYYYY - ZINGGGGG !!!!
How can you say that average daily return of 0.11% is just slightly better than 0.07%? 0.11/0.07 = 1.57, that means APPL has 57% bigger average daily return! And that is not slightly better at all, that is massively better and worth the extra volatility the stock has. Glad you are not portfolio manager ;)
Right?!?! Reminds me of the time my allowance for taking out the garbage went from $1 to $2 and I reveled in my 100% pay raise.