Result Detail – Glm

Back to Results Dashboard view Export CSV Export Excel Export PDF Download Full Report

Notebook

This result completed with warnings.

The data appear overdispersed relative to the Poisson model (Pearson phi = 9.47; values near 1 are expected). Coefficient standard errors are likely too small — refit with the Negative Binomial family.
The deviance goodness-of-fit test rejects adequate fit (p = 3.402e-280); residual structure may remain unexplained.

Test: glm — glm
Statistic: —
p-value: 0
Interpretation: significant Poisson GLM: GH_COUNT ~ ENV_SOILTN — smallest predictor p = 0
Value Column: GH_COUNT
AIC: 2948.86 pseudo-R² = 0.6544
Dispersion (Pearson φ): 9.474 — deviance GOF p = 3.402e-280
Observations: 207
Warnings: The data appear overdispersed relative to the Poisson model (Pearson phi = 9.47; values near 1 are expected). Coefficient standard errors are likely too small — refit with the Negative Binomial family.

The deviance goodness-of-fit test rejects adequate fit (p = 3.402e-280); residual structure may remain unexplained.

Term	Coef	SE	95% CI	z	p
const	4.5369	0.0196	[4.499, 4.575]	231.720	0
ENV_SOILTN	-5.4423	0.1192	[-5.676, -5.209]	-45.675	0

Coefficient estimates (95% CI) Download PNG

Observed vs fitted Download PNG

Residuals vs fitted Download PNG

Run Pathway Analysis

Method & Provenance

Method: Count GLM (Poisson / Negative Binomial)
Method key: glm (result type: statistical_test, result #182)
Engine: Python · statsmodels v0.14.6
Dataset(s): #700 gallery — catg_example_05_grasshopper_pest_pressure.csv
Status: completed
Permanent link: https://cat-g-omxpert.wishartlab.com/results/by-token/e5dac942c0614cd5bc9baa1859614534 Bookmark or share this URL — it does not depend on your session.
Created: 2026-09-21 02:15:12.174965

Reconstructed from the parameters stored with this result — represents the Python/R calls the analysis service made.

import statsmodels.api as sm

model = sm.GLM(y_counts, X_predictors, family=sm.families.Poisson())

fit = model.fit()