[R-sig-Geo] Parameter tuning of the bfastlite function

[Hugo Costa]

It's a pity if BFAST is not developed and maintained...
Hugo

Roger Bivand <Roger.Bivand using nhh.no> escreveu (sexta, 27/09/2024 à(s) 08:47):

> Could I suggest waiting until https://github.com/bfast2/bfast/issues/113
> which contains the same content has been attended to? Alternatively link to
> the list archives in the github issue to ensure that any responses to one
> channel do not get overlooked on the other channel (I've linked for now)?
> Emailing directly or github pinging the package developers may also help if
> this is urgent. Also consider reaching out to authors/maintainers of
> packages using bfast https://cran.r-project.org/package=bfast, and look
> (I'm sure you already have looked) among the many articles citing the
> underlying work. The package description mentions a forthcoming paper on
> BFAST Lite - this seems to be https://doi.org/10.3390/rs13163308. This
> isn't my field, but I think that bfast isn't being developed actively, and
> you may need to search broadly to try to resolve your issues, so looking at
> how other users handle this may help.
>
> Hope this helps,
>
>  Roger
>
> --
> Roger Bivand
> Emeritus Professor
> Norwegian School of Economics
> Postboks 3490 Ytre Sandviken, 5045 Bergen, Norway
> Roger.Bivand using nhh.no
>
> ________________________________________
> From: R-sig-Geo <r-sig-geo-bounces using r-project.org> on behalf of Nikolaos
> Tziokas <nikos.tziokas using gmail.com>
> Sent: 27 September 2024 02:19
> To: r-sig-geo
> Subject: [R-sig-Geo] Parameter tuning of the bfastlite function
>
> [You don't often get email from nikos.tziokas using gmail.com. Learn why this
> is important at https://aka.ms/LearnAboutSenderIdentification ]
>
> I am using the bfastlite() function from the BFAST package to run a
> time-series analysis. From the author's paper (BFAST Lite: A Lightweight
> Break Detection Method for Time Series Analysis) (table 2), I quote:
>
> "Needs parameter tuning to optimise performance, does not differentiate
> between breaks in seasonality and trend"
>
> So far, I was fine-tuning the model manually, that is, I was changing the
> parameters one by one, which is time-consuming. Does someone have a better
> solution regarding the fine-tuning of the model?
>
> To see which parameters of the model achieve the best results, I was
> checking the dates in the detected breakpoints (visual inspection). I am
> not sure if that method (visual inspection) is appropriate.
>
> I apologize if this question sound a bit vague, so let me expand a little
> bit. After running the bfastlite() using the default parameters (i.e., bp =
> bfastlite(datats)), we get a result. Is there a way to measure (something
> like rmse, or r-squared) how well the algorithm modeled the ts? What I
> basically mean is that if there is an index equivalent to let's say rmse
> when someone is running a linear regression. For example, what if the
> parameter breaks with BIC instead of LWZ detects more accurate the
> breakpoints (by visually inspecting the detected breakpoints)? Apart from
> the visual inspection, shouldn't be some other way to measure the
> performance of the model?
>
> Based on the above, is there a more efficient way to optimize the
> parameters of the model (based on some metric)? What do I mean by
> optimizing the parameters? I think with an example I can explain it better.
> When someone is tuning a random forest model, he/she can perform a full
> grid search to find the optimal parameters of the model (mtry, number of
> trees, etc) by searching all the possible combinations and for each
> combination he/she checks the rmse (or mse, r-squared). Is this what the
> authors of the paper meant when they said "Needs parameter tuning to
> optimise performance"? And if so, how did they do it?
>
> library(bfast)
>
> plot(simts) # stl object containing simulated NDVI time series
> datats <- ts(rowSums(simts$time.series))
>
> # sum of all the components (season,abrupt,remainder)
> tsp(datats) <- tsp(simts$time.series) # assign correct time series
> attributes
> plot(datats)
>
> # Detect breaks. default parameters
> bp = bfastlite(datats)
> plot(bp)
>
> # optimized model ??????
> bp_opt <- bfastlite()
>
> R 4.4.1, bfast 1.6.1, Windows 11.
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo using r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo using r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>

	[[alternative HTML version deleted]]