maker2qs: Get R2 values for linear regression of log catch anomalies vs...

In eeholmes/SardineForecast: Indian Oil Sardine Data, Code and References for Papers on Forecasting

Description

This function gets the R2 values for a linear regression of log catch anomaly ~ covariate anomalies at lags of 0 to 10. Covariates are turned into anomalies by removing year and qtr effect and then lagged and z-scored. Catch anomalies are logged and z-scored. Note: this is an old function and not used after I switched to using GAMs.

Usage

maker2qs(
  region = "Kerala",
  covname = "Precip",
  boxes = 1:14,
  rm.trend = TRUE,
  rm.season = TRUE
)

Arguments

region	region ("Kerala", "Karnataka", "Goa") in quotes.
covname	name of the covariate. Must be a colname in seio_covariates_qtr. If one of the covariates is from the boxes, then specify the part of the name in front of the number. For example "SST.".
boxes	for covariate from a box, the box number or numbers to use
rm.trend	remove year effect from catch and covariates
rm.season	remove season effect from catch and covariates

Value

A data.frame of R2's for each lag 0 to 10.

Examples

maker2qs(region="Kerala", covname="SST.", boxes=5)

eeholmes/SardineForecast documentation built on July 17, 2021, 2:56 a.m.