elemolotiv
  • Member for 4 years, 3 months
  • Last seen this week
3 answers
votes
18k views
Splitting Time Series Data into Train/Test/Validation Sets
13 votes

I think the most complete way to leverage your time-series data for training/validation/testing/prediction is this: Is the picture self explanatory? If not, please comment and I will add more text...

View answer
2 answers
votes
486 views
Bayesian inference - iterative updating with Bernoulli distribution (solved)
3 votes

thanks @Ben for insisting that I check my update process. It was broken indeed, I was not updating $P_{posterior}[\theta]$ correctly. So now I can proclaim that Bayesian inference with Bernoulli ...

View answer
2 answers
votes
585 views
How is sample mean divided by sample STD distributed for normal distributions?
1 votes

@Roman, here is my Python code, of @Ben's example above it produces a file "walks.tsv" with N realisations of a random walk with mean=μ and stdev=σ. for each realisation there is a row, with the ...

View answer
3 answers
votes
2k views
Cross validation : hyper-parameter tuning ? or model validation?
1 votes

Model Selection and Model Hyperparams Tuning are conceptually equivalent. If a model has a single Hyperparam, which in turn can have 3 possible values, it's like you have 3 different models. Very ...

View answer
3 answers
votes
138 views
Is this overfitting?
0 votes

Conceptually, the accuracy of your model is a function of these factors: complexity of the model number of samples in the training set variety of training samples (e.g. noisy samples VS repetitive ...

View answer