Questions tagged [statsmodels]
A Python module for exploration, testing, and estimation. Do not use this tag for general statistical modeling questions! Nb, questions only about the module itself, Python, or coding will likely be off topic.
393
questions
0
votes
0
answers
11
views
statsmodels.adfuller weird behaviour of usedlag value
I'm learning about time-series analysis and have two series on which I'm performing an Augmented Dickey-Fuller test in order to check for stationarity.
I'm trying to understand why I get very ...
- 101
0
votes
1
answer
23
views
Getting different AIC / BIC values for AR(2) estimation via AutoReg(2) vs ARIMA(2,0,0) through python statsmodels
I am trying to fit an AR(2) model to a data series claims_df['initial claims'] via statsmodels.tsa.ar_model.AutoReg and ...
- 1
0
votes
0
answers
12
views
Winning Bid probability [duplicate]
I had asked this question earlier, but I guess I was too late to respond to the questions. Hence reposting.
Top 10 win probabilities
I have a task at my hand to create some kind of model that can help ...
0
votes
1
answer
35
views
Doubts with cross correlation in Python's statsmodels
I have been trying to understand statsmodels.tsa.stattools.ccf, the cross correlation between two random vectors and I tried
...
- 271
0
votes
0
answers
13
views
what formula does python statsmodels library uses to compute the bp_stat value (the Q value for the Box Pierce test)
As far as I know, the $Q$ statistics is computed with the formula
$Q = N \sum_{j=1}^K \rho_j^2 $
I created an example where I compute $Q$ by hand and by using:\
...
- 101
0
votes
0
answers
23
views
+50
0
votes
0
answers
46
views
Top 10 win probabilities
I am a newbie here, hoping to find the right response for my question. I have a task at my hand to create some kind of model that can help me to determine what is the chance of winning a certain ...
0
votes
0
answers
12
views
Replicating a nested mixed effects model in R using python statsmodels
I am using this tutorial that is written in R to build a nested mixed effects model. The R code is as follows:
...
2
votes
1
answer
44
views
Linear model coefficient has a high p-value (insignificant) but the fitted value looks perfect
I have made an OLS model using statsmodels in python, in an attempt to model the response variable: energy cost per tonne. Note: ...
- 125
0
votes
0
answers
19
views
How to improve statsmodels SARIMAX forecast [duplicate]
I am trying to improve my statemodels SARIMAX model compared to the below naïve forecaster.
Naive forecast = y.mean() + ( X-X.mean() ) * y.std() / X.std()
where X is a series that has .75 ...
- 1
0
votes
0
answers
15
views
Is the likelihood of a discrete binomial variable same as it probability? Like in the case of tossing a coin lets say 12 times
I am working on a probability project where we first generate random variables for a given Binomial experiment and then we generate a PMF for 10 coin tosses using the list of random variables we ...
- 1
1
vote
0
answers
22
views
How do you calculate F-value for three groups, given only the mean and standard deviation in Python
I am learning about power analysis for different scenarios in Python, e.g. for three groups (assuming equal sample in each group). I've seen videos demo for SPSS and other softwares, where you simply ...
- 11
1
vote
0
answers
17
views
Modifying covariance structure in statsmodels mixedLM
I am using the following command to fit a linear mixed model for repeated measures (MMRM) in statsmodels:
...
- 21
0
votes
0
answers
32
views
Explained and Unexplained Effects don't add up to gap in Oaxaca-Blinder decomposition (statsmodels)
Here's a Python implementation of Oaxaca-Blinder decomposition analysis using statsmodels.
First, how do I interpret the negative value for Explained Effect here? ...
- 1
0
votes
0
answers
78
views
Python: Markov switching model out of sample forecasts
Is there a way to obtain out of sample forecasts for Markov switching models estimated via statsmodels (or any other package)?
https://www.statsmodels.org/dev/examples/notebooks/generated/...
- 123
0
votes
0
answers
35
views
Mixed model with nested fixed and nested random effects
This question is related to an earlier one that I've posted (but I must admit, that I was very unspecific in this thread, so I am making an effort to be more explicit here). I have a dataset that ...
- 151
0
votes
0
answers
43
views
Calculate proportion of variance explained in multiple linear regression with statsmodels
(Disclaimer: total regression noob!)
I have a multiple linear regression (one Y variable, several X variables) that I'm running using statsmodels.OLS. I would like ...
- 1
1
vote
1
answer
133
views
Offset in Poisson GLM with log link function where I have values equal to zero
I am trying to build a GLM model (poisson family) using python statsmodels package on train data. The data I have contains categorical values as exogenous variables and numerical values for my target (...
1
vote
0
answers
23
views
Holt's Winter Exponential Smoothing and LSTM? Some reference
Can you refer me a book, a site, or a PDF that can show a step by step guide for forecasting LSTM and Holt's winter Exponential Smoothing in a univariate data?
1
vote
1
answer
52
views
Why binary classification with highly imbalanced dataset output low probs?
I'm doing a frauds detection project which is using imbalance dataset(almost, 9:1).
And I using logistic, xgboost, and ligthgbm for Binary Classification.
When I predict my test set, I see all of ...
- 25
0
votes
0
answers
24
views
What are the best ranking models availble?
I am looking to compare data for example with sports teams, compare recent and historical results. I want to be able to compare any two teams when needed and also take into account a players database ...
0
votes
0
answers
134
views
Statsmodels seasonal decompose interpretation
I'm currently working with a time series that has a monthly frequency. It has 3 years of data, 2019-2021. When using the statsmodels.seasonal_decompose I got the following plots:
I would like to know ...
1
vote
1
answer
60
views
Implementing VAR predictions manually
I have trained a VAR model via statsmodels library in Python, and I am trying to implement the one-step ahead prediction manually without using the package.
Based ...
- 299
1
vote
0
answers
72
views
Correct use of repeated measures ANOVA
I have a repeated measures experiment with the following (simplified) dataframe.
...
- 1,427
0
votes
1
answer
43
views
Strange and very low results in Ljung-Box test
I have been reading some posts about the Ljung-Box test and I am applying it to some of my databases. However, I am not really understanding the outputs, I think I am doing something wrong. I have a ...
- 3
0
votes
0
answers
20
views
Estimating multiple parameters of a model in python [duplicate]
Wondering what's the most efficient/accurate way to estimate these parameters (a0, a1, a2, a3) with Python in the model:
...
- 101
1
vote
0
answers
22
views
Multiple testing correction in two dimensions
During one of our studies, we tried to investigate if there is a correlation between compound toxicity and specific protein expression levels. Both were measured in a considerable number of cell lines....
- 111
0
votes
0
answers
16
views
Why is the distribution of the FDR corrected P-values shifted so drastically towards high values?
I trying to understand why my distribution of the FDR-corrected P-values (Q-values) is shifted so drastically to the high values.
In this figure, on the left is the distribution of the (raw) values ...
user156469
2
votes
1
answer
139
views
Unclear which statistical test to use, t test or z test?
I have a scenario where I have an A/B test of a webpage. I'm trying to figure out using statistical analysis if more people spent time (in minutes, which is a float) on page B (the new page), than ...
- 45
1
vote
0
answers
30
views
Is it possible to do mixed ANOVA in statsmodels?
I'm analyzing an experiment where I have groups of people labeled y and o and all participants view three visualizations which ...
- 573
0
votes
1
answer
177
views
How do I interpret NaN values in statsmodels.stats.anova_lm result
I am trying to compare two models using statsmodels.stats.anova_lm. The output table I get is:
...
- 103
0
votes
0
answers
12
views
Minimal N days to reject sharpe ratio
A strategy assumes to have an annual sharp ratio of 8. Now backtest on N days trading data. If the N days return is negative, how many days does N need to reject the assumption that the annual sharp ...
- 885
0
votes
0
answers
99
views
What to do the following fit work using NegativeBinomial in statsmodel?
I am using following code to fit on given data but algorithm could not able to convergence. I believe this is due to high frequency of zero count. I am using both NegativeBinomialP and ...
- 61
1
vote
1
answer
21
views
SARIMAX exogenous coefs change with order of p and q
SARIMAX Statsmodels fits a regression with ARMA errors. So the model is:
yt=βxt+ut
ut=μ+ηt−1+ζt
Since p and q orders only affect the second equation, why do I get different results (coefs) of the ...
1
vote
1
answer
244
views
SARIMAX statsmodels model_fit.forecast predicts NaN values after 52
I have a dataset of bike renting between 2014 and 2020, and I want to use data between 2014 and 2018 to predict future usage, and to test it on 2019.
However when using predict, I get Nan values after ...
- 113
4
votes
0
answers
115
views
HAC standard errors: small-sample correction
The Python package statsmodels provides a use_correction option when computing HAC standard errors for an OLS model, which ...
- 252
0
votes
0
answers
32
views
PACF on Smooth Time Series
I have generated a heatmap of sorts from events occurring on certain dates. This heatmap is made as a Gaussian mixture with a fixed standard deviation of 50 days for each event. The heatmap looks like ...
- 125
1
vote
1
answer
105
views
Coefficient 0 for categories in statsmodels GLM
Is there a way to obtain the coefficient value 0 for the reference categories of categorical variables in statsmodels GLM.
...
0
votes
0
answers
96
views
Do you remove variables from a StatsModel logistic regression when you have a VIF of NaN?
I am building a logistic regression model to create a confusion matrix and decision tree. However, when checking for multicollinearity, I have a few variables that have a VIF of NaN. Do I remove them ...
0
votes
0
answers
42
views
Aggregating many correlations
I have measurements of covarying time-series. $X$ and $Y$ both vary in time, but I am interested in finding the correlation between $X$ and $Y$.
I have around 1500 measurements (individual timeseries) ...
0
votes
2
answers
106
views
Appropriate Regression Model for Proportions and Rate Data
I have a problem where my dependent variable is given as a click-through rate and thus bounded [0,1]. While I have the traffic for each sample (a combination of design factors) and could reconstruct a ...
- 1
0
votes
0
answers
31
views
Calculate Trends as a Percent per decade in a multilinear regression model
Am I calculating the trend value correctly? I am using a multilinear regression model on monthly mean data and several predictors. The model has an inflection point for the linear term. The answer I ...
- 1
3
votes
1
answer
126
views
Does this ARIMA model take seasonality into account?
I'm writing a tutorial on traditional time series forecasting models. One key issue with ARIMA models is that they cannot model seasonal data. So, I wanted to get some seasonal data and show that the ...
- 133
0
votes
0
answers
103
views
How to get incidence rate ratio of negative binomial regression with coefficients?
I'm running a NBR in Python's statsmodel pacakge. I want to see how a categorical variable influences website views. I got the model and coef and other stuff. Now I'm wondering how I interpret these ...
- 3
2
votes
1
answer
102
views
Prediction for time-series data gives constant values
I tried generating an AR process and checked whether it is predictable. After generating 5000 values of AR process, I put the first 4000 values as in-sample in statsmodels.tsa.ar_model.ar_select_order(...
- 23
1
vote
1
answer
163
views
Python for ARIMA model Monte Carlo? [closed]
I'm interested in fitting a time series with an ARIMA model in Python and then performing a Monte Carlo simulation to generate many possible future paths assuming the time series follows that model.
I'...
- 13
0
votes
0
answers
67
views
ARIMA(p,d,q) not converging with enforce_stationarity = False
I am using statsmodels package to fit ARIMA(p,d,q) model for several time-series. In a particular case, I get the following problem. When I fit the parameter using <...
- 1,511
2
votes
1
answer
218
views
How is confidence interval computed in statsmodels for ARIMA(p,d,q) parameters?
I am using statsmodel package for fitting ARIMA(p,d,q) model to a time series. My question is how exactly does this package estimate confidence intervals of the ...
- 1,511
1
vote
1
answer
109
views
Ordinal Regression: Python vs. SPSS
I need some help understanding the coefficients produced by Python (Statsmodels) for Ordinal Regression vs. SPSS. I ran the same exact data set in both SPSS and Python, but received different output ...
- 127
0
votes
0
answers
17
views
Forecasting Future Percentiles
I have been tasked to try and predict the percentile cutoffs that the NCQA - a committee that creates quality measures for Medicare programs. They use the rates from national programs for each ...
