Skip to main content
Akaike's Children's user avatar
Akaike's Children's user avatar
Akaike's Children's user avatar
Akaike's Children
  • Member for 5 years, 3 months
  • Last seen more than 4 years ago
60 votes
7 answers
8k views

Interview question: If correlation doesn't imply causation, how do you detect causation?

22 votes
5 answers
6k views

What is the intuitive meaning of having a linear relationship between the logs of two variables?

8 votes
1 answer
4k views

What is the difference between probabilistic forecasting and quantile forecasting?

6 votes
1 answer
277 views

How would someone use curves as an input to a supervised learning model?

5 votes
1 answer
4k views

Does a log transform always bring a distribution closer to normal?

5 votes
1 answer
2k views

Can VAR models handle time series with different lengths?

4 votes
1 answer
147 views

How does cross-validation work exactly?

3 votes
0 answers
877 views

Are Box-Cox and differencing redundant or complementary?

3 votes
1 answer
73 views

How do you forecast a time series that is inherently uncertain?

3 votes
0 answers
165 views

Do the assumptions for linear regression apply to AR(p) models?

2 votes
1 answer
192 views

Mapping one feature space to another for prediction purposes?

2 votes
0 answers
144 views

Is stationarity a requirement when using neural networks for time series forecasting?

2 votes
0 answers
2k views

What is the difference between using a Kalman filter and just recursively applying the Bayes rule as new data comes in?

2 votes
1 answer
5k views

How does feature selection work for non linear models?

2 votes
1 answer
216 views

Is a linear correlation between logs useful for making predictions with a regression model?

2 votes
1 answer
31 views

Distribution that acts like Poisson/NegBin for small means and like a Normal distribution for large means?

2 votes
1 answer
434 views

How is modeling the time series error/variance, e.g. ARCH or GARCH models, different from modeling time varying forecast intervals?

1 vote
0 answers
130 views

What is the resulting distribution of a data set that was originally normally distributed but has been quantized and had all negative values removed?

1 vote
1 answer
66 views

If I have a time series forecast density that is bi-modal, does that mean that my data is heteroscedastic?

1 vote
1 answer
514 views

Given the universal approximation theorem, why are LSTM better than feed forward neural networks at certain tasks?