All Questions
Tagged with statistics statsmodels
323
questions
0
votes
0
answers
7
views
Hausman test report [closed]
I've been working on a research project using a multi-level regression model, and I'm currently figuring out how to present the Hausman test results. I've seen some papers where authors mention doing ...
0
votes
0
answers
14
views
I try to understand F ditribution and relations to chi square and variances
Hello to everyone out there, I am trying to understand why F distribution and how it's related to chi-square. Why there are two formulas (with chi-square values devided by their corresponsing degrees ...
0
votes
0
answers
55
views
Getting rao score from MNLogit in statsmodels
I am doing a multinomial linear regression on a set of matrices which are essentially count tables each belonging to a class. My goal is to identify whether these tables are statistically similar or ...
0
votes
1
answer
28
views
Compute R Squared by Fixing Betas for Multi Linear Regression without Intercept
I want to fix the betas in multi linear regression based on some data I have, which leads to a RSquare value less than 0% and greater than 100 % based on the projection approach mentioned in ...
0
votes
1
answer
94
views
Is there a way to weigh datapoints when doing LOESS/LOWESS in Python
I would like to run a LOWESS function where different data points have different weights, but I don't see how I can pass weights to the lowess function. Here's some example code of using lowess ...
0
votes
0
answers
37
views
statsmodels qqplot doesn't match when using fit=True and scipy.stats.dist.fit
I was researching how quantile-quantile plots work and came across the following inconsistency:
I was trying to compare quantiles from uniformerly distributed sample against normal distribution ...
0
votes
0
answers
20
views
Is python statsmodel elastic net regression automatically standard deviation deflated?
I cannot find anywhere in internet for this really simple question so I have to ask here.
We know that applying regularization on norm of X you need to consider the standard deviation of your ...
0
votes
0
answers
24
views
Stats Model Power Calculation
The documentation is not very clear about the method for computing the number of samples estimated for a significant result, precisely the number of N. For example, does this N relate to one ...
1
vote
0
answers
47
views
Mixed effect model in python
Hi i have a certain task i want to solve:
i have this task:
For two months, participants played an app, in which they played 5 different
therapeutic games (TGs). At the beginning of each session, they ...
0
votes
0
answers
22
views
Statsmodels KernelReg standard errors
Is there a way to obtain prediction standard errors from statsmodel.nonparametric.kernel_regression.KernelReg function? I have seen on documentation that the fit attribute only returns point ...
0
votes
1
answer
75
views
Trying to gain intuition for curve-fitting function in scipy
I'd like to gain some intuition behind the scipy function scipy.optimize.curve_fit(). Below, I've written a simple version trying to minimise the sum of least squares function(the weights are all 1), ...
1
vote
1
answer
87
views
Getting a different kurtosis from numpy array method than from Summary
I have to extract information from the OLS statsmodel summary. While doing it, the Kurtosis results from the summary is different than the array method kurtosis().
Here is the code:
from sklearn....
0
votes
0
answers
62
views
LinAlgError("Singular matrix") with statsmodels logistic regression
I'm looking for associations between each predictor in a set of predictors x and each dependent variable in a set of outcomes y corrected for a set of covariates using univariate logistic regression ...
2
votes
1
answer
125
views
Replicating statsmodels.OLS mean squared error with scipy.stats.linregress
Is it possible to replicate the mean squared error of residuals from statsmodels.OLS using the stats.linregress? Please see my dataset with two regressions below. I can manually replicate the result ...
0
votes
0
answers
65
views
Is this code capturing both the forward and backward cross-correlation lags correctly?
I figured that statsmodel package's cross-correlation function sm.tsa.stattools.ccf (am testing two breathing signals recorded with different devices) only tests for forward lags and not for backward ...