TS_trend: Time series trend modelling
In ICRISAT-GEMS/SpaTemHTP: Processing and utilization of HTP data

Description Usage Arguments Details Value Author(s) References See Also Examples

Fit linear and logistic trends to G-BLUEs time series.

1	TS_trend(data, linear = TRUE, logistic = TRUE)

`data`	`numeric` `matrix` of dimension N_genotype x N_days containing the G-BLUEs time series. Such an object can be obtained with the function `SpaTemHTP_proc`.
`linear`	`Logical` value specifying if the linear trend should be fitted. Default = TRUE.
`logistic`	`Logical` value specifying if the logistic trend should be fitted. Default = TRUE.

The function fit a linear and/ or a logistic function to the genotype best linear unbiased estimates time series obtained with the function SpaTemHTP_proc. The logistic function is fitted using the function drm from package drc.

Return:

`lin_res`	`list` containing the parameters of the G-BLUEs TS linear trends.
`lin_R2`	`Vector` of r squared goodness of fit statistic for each G-BLUEs TS modeled with the linear trend.
`log_res`	`list` containing the parameters of the G-BLUEs TS linear trends.
`log_R2`	`Vector` of r squared goodness of fit statistic for each G-BLUEs TS modeled with the logistic trend.
`log_val`	`matrix` of predicted values according to the fitted logistic trend for each G-BLUEs TS.

Soumyashree Kar, Vincent Garin

Ritz, C., Baty, F., Streibig, J. C., Gerhard, D. (2015) Dose-Response Analysis Using R PLOS ONE, 10(12), e0146021

SpaTemHTP_proc

data(SG_PH_data)

SG_PH_data$col_f <- factor(SG_PH_data$col)
SG_PH_data$row_f <- factor(SG_PH_data$row)

SG_PH_data$rep <- factor(SG_PH_data$rep)
SG_PH_data$block <- factor(SG_PH_data$block)

exp_des_data = SG_PH_data[, c("row", "col", "row_f", "col_f","genotype",
"rep", "block")]

## Not run: 

G_BLUEs <- SpaTemHTP_proc(exp_des_data, pheno_data = SG_PH_data[, 6:8],
                          out_det = TRUE, miss_imp = TRUE, sp_adj = TRUE,
                          random = ~ rep +  rep:block + row_f + col_f,
                          plot = TRUE)
                          
G_TS_trend <- TS_trend(G_BLUEs)


## End(Not run)