User ManUtdBloke

19 votes

6 answers

3k views

When we plot data and then use nonlinear transformations in a regression model are we data-snooping?

asked Aug 22, 2020 at 13:01

11 votes

4 answers

9k views

Meaning of standard error of the coefficients in a regression model?

asked Sep 2, 2020 at 12:09

8 votes

1 answer

2k views

How to systematically choose which interactions to include in a multiple regression model?

asked Oct 21, 2020 at 14:42

7 votes

2 answers

254 views

If $X_n - \mu = O_p(a_n)$ does that imply that $X_n^{-1} - \mu^{-1} = O_p(a_n)$?

asked Feb 15, 2021 at 12:06

4 votes

1 answer

178 views

Does the variance in the estimates of the coefficients for standard linear regression decrease as $O\bigg(\frac{1}{n}\bigg)$?

asked Oct 8, 2020 at 10:15

3 votes

1 answer

154 views

Best practise for model selection when building predictive models?

asked Sep 7, 2020 at 9:02

3 votes

0 answers

692 views

Does the 'no serial correlation' condition for regression only make sense with respect to a sample and not the population?

asked Sep 25, 2020 at 13:45

3 votes

1 answer

716 views

Ridge regression not working for very simple dataset (yet exact same code works on another dataset)

asked Oct 6, 2020 at 13:56

3 votes

1 answer

915 views

Shouldn't the variance of the vector of regression coefficients decrease when we have larger sample sizes?

asked Oct 8, 2020 at 9:58

2 votes

0 answers

131 views

Unexpected results when applying convolution to variables created using rnorm?

asked Sep 16, 2020 at 9:48

2 votes

1 answer

240 views

Is the independence of this sequence of random variables not implicitly given when we define their probability distributions?

asked May 16, 2020 at 15:18

2 votes

1 answer

155 views

Why does the order of expectation plus the square root of order of variance equal the big $O_p$ rate of convergence in probability?

asked Mar 8, 2021 at 17:43

2 votes

1 answer

92 views

Is $\frac{\mu_X + O_p(n^{-1})}{\mu_Y + O_p(n^{-1})} = \frac{\mu_X}{\mu_Y} + O_p(n^{-1})$?

asked Dec 11, 2020 at 21:00

2 votes

1 answer

94 views

Peaks of estimated probability distribution are always lower than those of true distribution - why?

asked Feb 2, 2021 at 14:34

2 votes

1 answer

107 views

Sample size is $10^7$, what happens if we bootstrap with replacement using subsample of size $5000$?

asked Apr 5, 2021 at 13:51

1 vote

0 answers

69 views

Lots of modes of convergence for random variables but why is there nothing on convergence of probability density functions?

asked Oct 13, 2020 at 11:13

1 vote

2 answers

258 views

Approximating $E[g(\overline X_n)]$ and want to bound the remainder using some form of CLT or Berry-Essen Theorem

asked Mar 30, 2021 at 8:29

1 vote

2 answers

403 views

Conflicting p-values of regressors in simple regression vs multiple regression? [duplicate]

asked Oct 9, 2020 at 15:10

1 vote

1 answer

62 views

Deciding between linear and non-linear approach in a predictive model when the relationship 'looks linear'

asked Sep 7, 2020 at 14:56

1 vote

1 answer

3k views

If there any benefit to using ridge regression in a simple linear regression problem where the aim is prediction?

asked Sep 6, 2020 at 16:45

1 vote

1 answer

818 views

Is there a benefit to splitting the data by gender or age range when building predictive models?

asked Sep 7, 2020 at 8:28

1 vote

2 answers

162 views

If the $\varepsilon$ in $Y = \beta_0 + \beta_1 X + \varepsilon$ does not represent measurement error in $Y$, then what does it represent?

asked Sep 18, 2020 at 11:39

1 vote

1 answer

244 views

Is there any need for regularization in an overdetermined multiple regression problerm?

asked Aug 18, 2020 at 12:22

1 vote

1 answer

71 views

Is there an error Section 6.6.2 of the book An Introduction to Statistical Learning?

asked Oct 6, 2020 at 16:02

0 votes

0 answers

116 views

Bounding the norm of the difference between two related probability densities

asked Oct 6, 2020 at 12:13

0 votes

0 answers

135 views

How to mathematically describe convergence of sample density to population density?

asked Sep 12, 2020 at 13:09

0 votes

0 answers

344 views

Is a non-parametric density estimation required for a bimodal distribution?

asked Aug 12, 2020 at 10:37

0 votes

0 answers

34 views

Bounding the difference between the PDF of $\hat\beta\sim\mathcal{N}(\beta,\sigma^2)$ and a Dirac delta $\delta_\beta$ in terms of the variance?

asked Oct 13, 2020 at 9:25

0 votes

0 answers

33 views

Drop or impute predictor that is missing around 70% of values, but which is known to be highly relevant?

asked Oct 9, 2020 at 13:20

0 votes

0 answers

75 views

Is the inverse of the sample variance integrable?

asked Mar 17, 2021 at 20:08

Stack Exchange Network

32 Questions

When we plot data and then use nonlinear transformations in a regression model are we data-snooping?

Meaning of standard error of the coefficients in a regression model?

How to systematically choose which interactions to include in a multiple regression model?

If $X_n - \mu = O_p(a_n)$ does that imply that $X_n^{-1} - \mu^{-1} = O_p(a_n)$?

Does the variance in the estimates of the coefficients for standard linear regression decrease as $O\bigg(\frac{1}{n}\bigg)$?

Best practise for model selection when building predictive models?

Does the 'no serial correlation' condition for regression only make sense with respect to a sample and not the population?

Ridge regression not working for very simple dataset (yet exact same code works on another dataset)

Shouldn't the variance of the vector of regression coefficients decrease when we have larger sample sizes?

Unexpected results when applying convolution to variables created using rnorm?

Is the independence of this sequence of random variables not implicitly given when we define their probability distributions?

Why does the order of expectation plus the square root of order of variance equal the big $O_p$ rate of convergence in probability?

Is $\frac{\mu_X + O_p(n^{-1})}{\mu_Y + O_p(n^{-1})} = \frac{\mu_X}{\mu_Y} + O_p(n^{-1})$?

Peaks of estimated probability distribution are always lower than those of true distribution - why?

Sample size is $10^7$, what happens if we bootstrap with replacement using subsample of size $5000$?

Lots of modes of convergence for random variables but why is there nothing on convergence of probability density functions?

Approximating $E[g(\overline X_n)]$ and want to bound the remainder using some form of CLT or Berry-Essen Theorem

Conflicting p-values of regressors in simple regression vs multiple regression? [duplicate]

Deciding between linear and non-linear approach in a predictive model when the relationship 'looks linear'

If there any benefit to using ridge regression in a simple linear regression problem where the aim is prediction?

Is there a benefit to splitting the data by gender or age range when building predictive models?

If the $\varepsilon$ in $Y = \beta_0 + \beta_1 X + \varepsilon$ does not represent measurement error in $Y$, then what does it represent?

Is there any need for regularization in an overdetermined multiple regression problerm?

Is there an error Section 6.6.2 of the book An Introduction to Statistical Learning?

Bounding the norm of the difference between two related probability densities

How to mathematically describe convergence of sample density to population density?

Is a non-parametric density estimation required for a bimodal distribution?

Bounding the difference between the PDF of $\hat\beta\sim\mathcal{N}(\beta,\sigma^2)$ and a Dirac delta $\delta_\beta$ in terms of the variance?

Drop or impute predictor that is missing around 70% of values, but which is known to be highly relevant?

Is the inverse of the sample variance integrable?