fit_simple_polynomial: Fitting Simple Polynomial Models
In pieterbeck/CanHeMonR: Tools for canopy health monitoring through image analysis

Description Usage Arguments Details Value Note

View source: R/fit_function_to_canopy_health_index.r

Fits a simple polynomial of the form y ~ scale*(x^rate)+interc, using nls, designed to be fit to a time series of a canopy health index that increases(?) in case of anomaly

1 2	fit_simple_polynomial(data, scale. = 2.8, rate. = 1, interc. = 0, plott = F, min.n.time. = 3)

`data`	Data frame with two variables: im.dates, the independent variable, and DN.c, the dependent one (e.g. a VI or reflectance)
`scale.`	The starting value for the scale parameter in the nls estimation of the polynomial. It determines how high the values rise. Default is 2.8.
`rate.`	The starting value for the rate parameter in the nls estimation of the polynomial. rate. determines how abrupt the polynomial increases. rate. = 1 generates a linear increase. Default is 1.
`interc.`	The starting value for the intercept parameter in the nls estimation of the polynomial. Default is 1.
`plott`	Logical. Should results be plotted. Default is FALSE.
`min.n.time.`	The minimum nr of unique time observations for the model to run

The current formulation of the model, combined with a scaling of the tim ethat makes it start at 0, forces the model through VI==interc at the first time observation the rate determines how abrupt the increase/decrease is. Rate=1 generates a linear increase Note that when the rate is 0, the model fails

An nls model

called by decline_trees_RS_analysis.r

pieterbeck/CanHeMonR documentation built on May 25, 2019, 7:11 a.m.