Cormack-Jolly-Seber model (Sciuridae)

Knowledge graph centered on Cormack-Jolly-Seber model (Sciuridae) with 46 nodes and 156 connections. Top connected: Rocky Mountain Biological Laboratory, Gothic, Marmota, Yellow-bellied Marmot, Crested Butte.

Method synopsis

Semi-parametric survival analysis using Cox proportional hazard models to test effects of maternal, social, and environmental covariates on annual survival probability while allowing baseline hazard to vary with time.

Synthesized from method descriptions across 2 papers using this protocol.

Procedure from a recent peer-reviewed implementation

Steps below were extracted from the most recent peer-reviewed implementation of this protocol in the corpus — Maternal survival costs in an asocial mammal (2022), Ecology and Evolution. Implementations in other papers (listed below) may differ.

1. Cox proportional hazard model analysis
   Cox proportional hazard models (CPH; Cox, 1972) were used to assess the proportional hazard assumptions of the non-parametric Kaplan-Meier model (1958), to estimate annual survival as a function of the covariates of interest. Cox proportional hazard models are semi-parametric and allow the mortality hazard to fluctuate with time while measuring the effects of covariates on either age- or time-specific survival. An individual's contribution to the mortality hazard consists of an entry time, an exit time, and whether the animal survived the interval. The effect of any given covariate at time t was measured on the likelihood of an individual surviving the interval from t to t+1 (i.e., from spring of a given year to the following spring), as opposed to surviving the animal's entire lifespan, which would range across multiple years in some cases. The function 'cox.zph' was used to assess the proportional hazard assumptions of (1) to ensure covariates weighed evenly on the mortality hazard.
   Quantities: 249 annual observations, 131 mortality events, tested multiple parameterizations for each variable typeDuration: Analysis period covered 1995-2020 dataConditions: Statistical analysis using complete dataset
   Equipment: RStudio (R Core Team, 2019), survival package (Therneau & Grambsch, 2000), survminer package (Kassambara & Kassambara, 2020), AICcmodavg package (Mazerolle & Mazerolle, 2017)

Typical Equipment

• RStudio (R Core Team, 2019)
• survival package (Therneau & Grambsch, 2000)
• survminer package (Kassambara & Kassambara, 2020)
• AICcmodavg package (Mazerolle & Mazerolle, 2017)

Output Measurements

• hazard ratios
• confidence intervals
• p-values
• survival probabilities

Papers Using This Protocol (2)

Species Studied With This Protocol

Places Where Used