Brief introduction of Python statistical regression in Python mathematical modeling library
- 2021-12-09 09:28:43
- OfStack
1. About StatsModels
statsmodels (http://www. statsmodels. org) is an Python library for fitting various statistical models, performing statistical tests, and data exploration and visualization.
2. Documentation
The latest version of the document is located at:
https://www.statsmodels.org/stable/
3. Main functions
1. Linear regression model:
Ordinary minimum 2 multiplication Generalized least 2 multiplication Weighted least 2 multiplication Minimum 2 multiplication with autoregressive error Quantile regression Recursive minimum 2 multiplication2. Mixed linear models with mixed effects and variance components
3. glm: Exponential family distribution of generalized linear model supporting all one parameter
Bayesian hybrid glm of term 4.2 and poisson
5. gee: Generalized estimation equation for unidirectional clustering or longitudinal data
6. Discrete model:
logit and probit Polynomial logit (mnlogit) Poisson and Generalized Poisson Regression Negative binomial regression Zero expansion counting model7. rlm: Robust linear model supporting multiple m estimates.
8. Time series analysis: Time series analysis model
Complete state space modeling framework Seasonal arima and arimax models Varma and Varmax models Dynamic factor model Unobserved component model Markov switching model (MSAR), also known as hidden Markov model (HMM) Univariate time series analysis: ar, arima Vector autoregressive model, var and structure var Vector error correction model, vecm Exponential smoothing, Holtwinters Hypothesis testing of time series: unit root, cointegration, etc. Descriptive statistics and process model of time series analysis9. Survival analysis:
Proportional hazard regression (cox model)
Survivor function estimation (kaplan-meier)
Estimation of cumulative correlation function
10. Multivariate:
Principal Component Analysis of Missing Data Rotation factor analysis Manova Canonical correlation11. Nonparametric statistics: univariate and multivariate kernel density estimates
12. Datasets: Datasets for example and test
13. Statistics: Extensive statistical tests
Diagnostic and specification testing Goodness of fit and normality test Multiple test function Various additional statistical tests14. Mouse interpolation, sequential statistical regression and Gaussian interpolation
15. Intermediary analysis
16. Graphics include mapping functions for visualizing analytical data and model results
17. Input/Output
A tool for reading stata. dta files, but pandas has a newer version Table outputs are ascii, latex, and html18. Sandbox: statsmodels contains a sandbox folder containing development and tests that are not considered "production ready".
Estimator of generalized moment method (gmm) Nuclear regression Various extensions of scipy. stats. distributions Panel data model Information theory measure
4. Acquisition and Installation
pip3 install --upgrade statsmodel -i https://pypi.tsinghua.edu.cn/simple
The above is the Python mathematical modeling library StatsModels statistical regression introduction of the first detailed content, more about the mathematical modeling library StatsModels statistical regression information please pay attention to other related articles on this site!