Anders Munk-Nielsen

Пікірлер

@ruanvermeulen7594 23 сағат бұрын

Wow! Speak of clear straightforward!

@AssafZipori-g4d Ай бұрын

Thank you for this great video!!! Can you please provide a reference for this procedure?

@tallyskalynkafeldens1753 3 ай бұрын

Great explanation! Thank you very much!

@areejareej281 5 ай бұрын

Is kernel regression is another name of kernel ridge regression?

@aidansloyan1373 7 ай бұрын

Really awesome !

@mokus603 8 ай бұрын

Thank you for your time and effort! This is such a good video and while you're talking about the regression part, you didnt skip explaining parts of the code with examples (numpy vectors, shapes, etc.)

@iyasudiro7371 10 ай бұрын

❤❤❤❤❤

@yurisantos9251 11 ай бұрын

Foud iddle!

@maxinavarrette Жыл бұрын

Hi thanks for the video. I am wondering if you could offer a more rigorous explanation for why one needs to include h_outer & h_inner for the 2nd derivative? I understand linear and quadratic approximations quite well, but i am having trouble translating my knowledge to python

@superpronker 11 ай бұрын

I can't find a good source as it stands right now, so the best I Cana do is to point you to the arguments for the first derivative (en.wikipedia.org/wiki/Numerical_differentiation) and then to note that: taking a 2nd derivative is like taking a 1st derivative twice. Suppose the original function has (almost) no numerical noise; the 1st derivative is numerical, so it will have a fair bit more noise. Therefore, when we are choosing the step size for the 2nd derivative, the tradeoff between "numerical precision / roundoff errors" versus "analytical precision / finite difference error" is affected - we have to lean more to the side of avoiding numerical errors, which we do by taking a larger (coarser) step size. Hope this helps.

@user-fe6vi3yg8i Жыл бұрын

This is fantastic! Thanks so much for the detailed explanations! Quick follow up query. The method at the beginning, compute_full_matrix, doesn't seem to be a part of the bimatrix source. Could you kindly confirm where to find that implementation? When I checked out William Spaniels video, not sure if I missed it, but got no reference to the python code of this method there...

@superpronker Жыл бұрын

Thanks, glad you enjoyed. The function is inside this notebook towards the top: github.com/GamEconCph/Lectures-2021/blob/main/Bayesian%20Games/BNE.ipynb

@user-fe6vi3yg8i Жыл бұрын

@@superpronker Awesome! Once again, thanks so much for your time and efforts in teaching this! I believe I hadn't toggled to the Code's tab in the notebook at the time, and missed it. The default in github is to show the Preview tab, trivial, but easy to miss. Still new to Jupyter, so thanks for hand-holding!

@leoWorldBestGamer Жыл бұрын

GREAT VIDEOS!!!!!!!

@Justin-zw1hx Жыл бұрын

excellent explanation

@iamZANIX Жыл бұрын

Thanks

@skipperkongen Жыл бұрын

Rigtigt god video, tak!