Predict function accuracy to increase

[ron4]

I have a working Predict function.Now it needs a much higher accuracy as the prediction is not accurate enough.Its not even close.
Is there a package or code anyone knows how to increase accuracy of the below predict function?

reprex:

# import data
setwd("C:/Users/jenny/Desktop/R Studio")
sales <- read_excel("C:file)")

sales <- ts(sales, frequency=3, start=c(2019,1))
predict(ts(sales))

^{Created on 2020-06-18 by the reprex package (v0.3.0)}

[williaml]

The reprex isn't of any use as it is reading from the file on your computer.

It looks like you are trying to predict the next few values of a time series though. Try the fable package.

Forecasting Models for Tidy Time Series

[ron4]

Hi, i will test it for sure.Thank you William.

[ron4]

Hi,i tried to test it,but i cant get the code correct.
Does it show a prediction with Alpha, Beta and Gamma?
Cheers

[williaml]

It would depend on the model you are using. There is an example with an ETS model with the alpha, beta and gamma smoothing parameters:

https://fable.tidyverts.org/articles/fable.html