Quantile regression explained: Estimating conditional quantiles (Excel)

Quantile regression is an intuitive and flexible data analysis tool first proposed by Roger Koenker in 1978. Unlike linear regression that estimates conditional mean, quantile regression can be used to estimate the conditional median or any conditional quantile or percentile of your data. Today we are learning how to apply quantile regression in Excel, estimate its loss function, covariance matrix, and standard errors, and discuss its applications in finance and risk management.
Don't forget to subscribe to NEDL and give this video a thumbs up for more videos in Econometrics!
Please consider supporting NEDL on Patreon: / nedleducation

[@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

[@lmagz84]

I have seen your videos and they are really helpful thanks alot. But I want to measure VaR of banking system Conditional on bank i being at its VaR level.CoVaR by Tobias Adrian and Markus K. Brunnermeier 2011 , 2016. They performed it using quantile regression. If you don't mind I can contact you via email, for more details

[@NEDLeducation]

@@lmagz84 Hi again Feryal, and thanks for the clarification and the follow-up question! The Adrian and Brunnermeier approach is relatively straightforward - they use a quantile regression to estimate how each of the banks' performance impacts the overall performance of the banking sector at the lower quantiles (0.01 for 99% CoVaR and 0.05 for 95% CoVaR). They use data on banking industry index returns and match it with quarterly banking fundamentals (such as assets, liabilities, leverage, etc.)

[@lmagz84]

@@NEDLeducation Thanks alot, for ur understandable way. I am trying to run quantile Eviews. But I think in addition its Conditional on bank being in distress bring at its VaR level, right? I only have daily returns for banks and system, how to express the bank being in distress in quantile regression?

[@dumbwaiter]

Great video. QR is actually very handy for forecasting demand. If you have a pivot table showing your sales per M, Q or Y, you can then use excel's built in forecast models, and perhaps add one or two autoregressive models (arima1,0,0 or arima1,1,0) yourself into the pot and then make a quantile regression on the forecast of each model (Ie., bundling up different model predictions), you may get very good predictions. Gonna try it. Thanks

[@willmccubbin3240]

Hello Savva, I hope you are well. It would be very interesting if you could consider modelling the development of Green Investments and do a deeper dive into the current performance compared to various other asset classes. Thank you for all the videos you have created. Many thanks William McCubbin

[@NEDLeducation]

Hi Will, happy to hear from you and see you are still checking out the channel :) As for your suggestion, this is really ambitious and might involve writing a bunch of papers rather than recording a couple of KZbin tutorials, but I am definitely considering doing something along the lines of ESG rating, ESG risks, and ESG stock performance in the future. Stay tuned for more content :)

[@andrewlin4587]

Really nice video, well explained. Thank you very much.

[@shreyashdeshmukh1470]

Thanks Bro, its been a great help to me

[@corradoforza]

Thank you so much, as always you make all very clear and easy to understand! Video suggestion: principal component analysis 🚀

[@NEDLeducation]

Hi again Corrado, and glad you liked the video! I might consider doing one on PCA quite soon!

[@Maria-tn4cn]

you are doing great

[@sangeethamathews2136]

Very useful

[@nicolaspatassi3877]

thank you so much for your videos which are definitely usefull. I would really appreciate whether you might publish something related to liquidity risk measurement for bonds portfolio and equities ptf with a regulation obligation point of view : whatever the mean to get results (daily volumes, participation rate, anithing else..), the result is a liquidity ladder with liquidity bucket to assess time scale to sell all assets. Thank you again and continue straightforward your job!

[@NEDLeducation]

Hi Nicolas, and glad you are enjoying the channel! As for your suggestion, there is a video on liquidity risk for investment management already in the pipeline due to be released shortly, so stay tuned! For regulatory aspects of liquidity risk, I do cover Basel metrics and ratios such as LCR and NSFR please check these out if you are interested: kzbin.info/www/bejne/h6W6d2ecgKyMldE and kzbin.info/www/bejne/gHzHmIuFm52tsNk. Hope it helps!

[@nicolaspatassi3877]

@@NEDLeducation nice! I saw those videos you are talking about on banks regulation. I am more focus on investment management risks calculation, and glad to heard you schedule a video within that frame. Really good job!

[@lmagz84]

Please I need help to measure Co Value at Risk as a systemic risk measure for banking system, can you help me and you can price your service

[@NEDLeducation]

Hi Feryal, and thanks for your question. If you mean Conditional Value at Risk (CVaR, and also known as Expected Shortfall), I have got a couple of videos on that so check these out: kzbin.info/www/bejne/pabGk5yffbCpfc0 and kzbin.info/www/bejne/gIbSlIZ8mpJ3ipI.

[@ruchitadmello93]

@@NEDLeducation Even I need help to measure CVaR as a systemic risk measure for banking system. Can you provide me your email address as I have few queries

[@Im-Assmaa]

Hi, thank you so much, this is so detailed and beautifully explained. I have a question. How can I calculate Quantiles for a specific p, using the Rankit -Cleveland definition?

[@saminrasi]

Thanks a lot for this video! Since the Wald test is not valid in quantile regression, how can we test for the equality of coefficients? (like in causality tests)

[@yulinliu850]

👍

[@deniswolf1846]

Савва, добрый день, я Вам на почту писал по поводу курсов, подскажите пожалуйста, Вам удалось прочитать? Заранее большое спасибо!

[@NEDLeducation]

Добрый день, конечно, спасибо за идею. Планирую выпускать больше более базовых видео по финансовому моделированию в Python. Насчет более структурированных курсов - возможно в более отдаленном будущем.

[@deniswolf1846]

@@NEDLeducation Круто! Огромное Вам спасибо!

[@capiramirez]

Just great explanation.

[@dr.kingschultz]

Very good videos!