OHLC volatility (Part 1) - Parkinson and Garman Klass (Excel)
HTML-код
- Опубликовано: 18 май 2020
- Is the standard deviation of close-on-close stock return the best measure of volatility? Some might argue it is not as it misses important information contained in opening, high, and low prices and disregards the risk of intraday trading. Today, we are discussing and calculating some of the easiest measures of OHLC volatility, namely, Parkinson and Garman and Klass estimators that has been developed in the 1980s.
Don't forget to subscribe to NEDL and give this video a thumbs up for more videos in Finance!
Please consider supporting NEDL on Patreon: / nedleducation
You can find the spreadsheets for this video and some additional materials here: drive.google.com/drive/folders/1sP40IW0p0w5IETCgo464uhDFfdyR6rh7
Please consider supporting NEDL on Patreon: www.patreon.com/NEDLeducation
Придётся смотреть все Ваши видео ) спасибо
literally this channel is so underrated. loved your content. thank you sir.
Hi Rahul, thank you for the feedback! Glad you liked the videos :)
So true, this video is solid gold.
Number of views is obsviously uncorrelated to this premium quality video.
Thank you @NEDL
beautiful content
THANK YOU VERY MUCH! Your videos are very clear and helpful.
what a teasing at the end of the video ... love it ! ^^
ty my good man, now i have correlation to nail and then im ready to submit me assignment in 5 hours
Great Video!
As usual, a very good presentation.
Hi Ronald, thanks for your feedback :)
Good channel
Отлично! Спасибо! Как раз то что искал.
Было бы супер если бы и файл выложили :)
Виталий, спасибо большое за фидбэк! Напишите на s.shanaev@northumbria.ac.uk, вышлю файл. :)
Thank you for the video! Quick question: if I want to use German-Klass to estimate daily volatility (as an alternative of a sum of squared intraday returns), do I just omit the summation and 1/n division?
how can I use OHLC data to warn of upcoming volatility ? and for tops and bottoms?
Can you suggest some books for further reading into these type of topics please?
Nice content. Can we talk about high frequency sampled volatility too?
Thanks for your feedback :) We were planning to do a video on variance ratio test for some time now actually, we might cover this concept there. However, if we think about high-frequency volatility in simple terms, in is just close/close volatility but calculated over different time horizons. And then it mainly becomes a volatility scaling issue.
Interesting topic and great presentation. Would you share that spreadsheet? Would be great to be able to build upon it. Thanks in advance
Hi Massimo, and many thanks for your feedback! Just drop me an email on s.shanaev@northumbria.ac.uk and I will share the spreadsheet with you :)
can you please advise if you can FORECAST prices using OHLC data? Many trader in India claim that they FORECAST the trend using OHLC data in excel but no one has shared it.
Hi Sachin, and many thanks for the question! Generally, it is a consensus in finance that price forecasts based on OHLC are not allowing one to develop trading strategies that perform better than the market, particularly for liquid assets. This is one of the consequences of the efficient market hypothesis that is not always true for financial markets, but mostly true for liquid markets such traders operate on.
What an outstanding presentation, Sir! I have a question, what is the main difference between Parkinson and Garman-Klass volatility besides the use of data in the calculation?
Hi, and thanks so much for such kind words! Long story short, Garman-Klass can be considered a generalisation of Parkinson that does not underestimate volatility that severely, however they both do not account for price jumps during non-trading hours and drift (non-zero mean return).
@@NEDLeducation oh I see.. Another question, if we try to calculate the Parkinson volatility estimation of each day, it means the formulation is "SQRT(LN of high divided by low of that day)...." am I right, Sir?
@@naufalwee Yes, that is correct.
@@NEDLeducation ahh, thank you very much, Sir! Appreciate it a lot!🙌🏻
Савва, у вас серьезная ошибка в формуле для эстиматора Гарман-Класса, должно быть -(2*ln(2)-1)*(ln(Close/Open)), то есть "-", а не "+".
Hi NEDL. Please checking in formula of GK. I see add( +) or minus (-) before (2ln(2)-1)ln^2(C/O)). Some article see difference
Source: + core.ac.uk/download/pdf/52391988.pdf ( formula 3)
+ dynamiproject.files.wordpress.com/2016/01/measuring_historic_volatility.pdf ( formula of Garman Klass Page 9)
look like mistake and I don't know about that. thank for all !