tsfngm (version 0.1.0)

FFNGM: Fitting and Forecasting using Nonlinear Growth Models

Description

The FFNGM function fits nonlinear growth models to time series data and computes the h step ahead forecast values.

Usage

FFNGM (x, t, model=c("Monomolecular", "Logistic", "Gompertz"), k, y, r, h)

Arguments

x

a univariate time series data.

t

a numeric vector containing time points.

model

<U+201C>Monomolecular<U+201D> or <U+201C>Logistic<U+201D> or <U+201C>Gompertz<U+201D>.

k

Initial estimate of carrying capacity (maximum limit of the considered time series data).

y

Initial estimate of starting value of the considered time series data.

r

Initial estimate of growth rate.

h

The forecast horizon.

Value

modelsummary

Summary of the fitted model

fitted.values

Fitted values of the model

MAE

Mean Absolute Error (MAE) of the fitted model

MAPE

Mean Absolute Percentage Error (MAPE) of the fitted model

MSE

Mean Square Error (MSE) of fitted the model

RMSE

Root Mean Square Error (RMSE) of the fitted model

forecasted.values

h step ahead forecasted values of the fitted Model

Details

Using the nonlinear least squares method, this function estimates the parameters of nonlinear growth models for time series data. This function returns the fitted model summary, as well as the model's fitted values and various evaluation criteria. This function also returns the fitted model's h step ahead forecasted values.

References

Pal, S. and Mazumder, D. (2015). Forecasting groundnut production of India using nonlinear growth models. Journal Crop and Weed, 11, 67-70.

Seber, G. A.F. and Wild, C. J. 2003. Nonlinear Regression, 2, New York: John Wiley.

See Also

nls

Examples

# NOT RUN {
t<-c(0:12)
x<-c(57.97,66.02, 72.62, 77.87, 81.95, 85.07, 87.43, 89.20, 90.52, 91.50, 92.22, 92.75, 93.1)
FFNGM(x,t,"Gompertz",94, 55, 0.1, 3)

# }