================================================================================================ also, would “try… except… else” be able to skip and move on to next forecast?

Hello everybody. I am testing Nixtla for some hierarchical forecasts and I am getting the following error. Any one know what I can do? Thanks

Hello again. The previous error is no longer happening. I restarted the runtime and it started working. Got very excited by the fact that there is a forecast. However I then had an error in the reconciliation. It says the array cannot contain infs of Nans (see screenshot). I checked the three inputs to see if there were any. If my check is correct (see screenshot) there are none. Any ideas? Thank you for all your help.

Thanks for reporting the error. Let us have a look and come back to you.

Hi, I am using the aggregate function to generate the hierarchical data (actuals). However, I don't need to aggregate my forecasts since I generate forecasts at each level. The notebook examples use the aggregate function. Does anyone have an example of how my data should look in order to use the HierarchicalLoad function? When I run this, it triple counts everything since I have forecasts at all 3 levels:

Y_fitted_df, S_fitted, tags_fitted = aggregate(Y_fitted_df, hiers)

When I run this, I get an error:

Y_fitted_df, S_fitted, tags_fitted = HierarchicalData.load(Y_fitted2_df)

TypeError: load() missing 1 required positional argument: 'group'

Hello! Question in regards to neuralforecast. I'm training a GMM_TFT model. I noticed one does not pass a validation dataset to the trainer via the fit method. How then does one use an early stopping callback based on validation loss?

Hello! I have a question regarding model experiment. I am trying to log statsforecast models in databricks ML Flow environment. Is there any reference how I can do it? I had the code as below but always got the error of "KeyError: 'artifact_uri'"

Hi, I am using statsforecast with multiple models (Auto ARIMA, ETS, Random Walk) and I want to get insample forecast using forecast_fitted_values. But it seems that this function cannot take multiple models at the same time. Is there any best practice other than splitting 3 models into multiple 3 fit&forecasts?

not sure what the roadmap looks like for stats / mlforecast packages but we’ve had some surprising results with LEAR using this implementation https://github.com/jeslago/epftoolbox/blob/master/epftoolbox/models/_lear.py

If I use fit predict interface on 100 timeseries, what are the saved weights if I pickle the StatsForecast object?

Do you know if the

forecast

function in the utils can take exogenous regressors? Code snippet in thread

Hello, I'm getting this error for a very simple dataset

RemoteTraceback Traceback (most recent call last) RemoteTraceback: """ Traceback (most recent call last): File "/Users/jose.bordon/opt/anaconda3/lib/python3.9/multiprocessing/pool.py", line 125, in worker result = (True, func(*args, **kwds)) File "/Users/jose.bordon/environments/project1_env/lib/python3.9/site-packages/statsforecast/core.py", line 159, in forecast raise error File "/Users/jose.bordon/environments/project1_env/lib/python3.9/site-packages/statsforecast/core.py", line 154, in forecast res_i = model.forecast(h=h, y=y_train, X=X_train, X_future=X_f, fitted=fitted, **kwargs) File "/Users/jose.bordon/environments/project1_env/lib/python3.9/site-packages/statsforecast/models.py", line 303, in forecast mod = auto_arima_f( File "/Users/jose.bordon/environments/project1_env/lib/python3.9/site-packages/statsforecast/arima.py", line 1655, in auto_arima_f fit = Arima(x, order=(0, 0, 0), include_mean=False) File "/Users/jose.bordon/environments/project1_env/lib/python3.9/site-packages/statsforecast/arima.py", line 1316, in Arima tmp['bic'] = tmp['aic'] + npar*(math.log(nstar) - 2) ValueError: math domain error

I checked that there are no nulls in my data set

sf = StatsForecast(df=df_agg2, models=[AutoARIMA()], freq='W', n_jobs = -1) forecasts = sf.forecast(7) forecasts.head()

picture is df_agg2

Hello team, first thanks for all the amazing (and publicly available) work. Quick questions about statsforecast (potentially the other libraries) I can’t quite figure out from the documentation: Once I trained my models, what is the best way to make predictions on the test set for 3 day horizons as opposed to the whole test set starting from the last training date? • Would one have to apply the .forecast method iteratively one day at a time setting df (or y) to the data up to that day and h to 3? • Should one reuse the cross_validation method setting the df and the intpus_size/test_size/n_windows smartly to do this more efficiently?

We did a new release in

HierarchicalForecast

, show some love community ❤️ https://twitter.com/fede_gr/status/1592621393975586817

TCN work looks great btw, tossing this here in case there’s interest in it: https://lnkd.in/eBcGjh3V

We played around with neural ODEs a bit earlier this year but found them very slow to train and quite unstable, really excited about the new research direction!

Currently trying to migrate from

fable

to statsforecast in our production environment. However, I’m running into an issue with replication. Is it possible to

boxcox

transform prior to the forecasting step and back transform to produce the forecast mean rather than median? I’ve tried using the

scipy

python package, but it’s producing lower predictions than the R implementation. Any ideas / thoughts? Examples: ETS ->

fable::ETS(fabletools::box_cox(qty + 1, lambda = 0.3), opt_crit = "mae")

ARIMA ->

fable::ARIMA((fabletools::box_cox(qty + 1, lambda = 0.4)) ~ PDQ(period = 13), stepwise = TRUE)

THETA ->

THETA(fabletools::box_cox(qty + 1, lambda = 0.1) ~ season(method = "additive"))

I think this will be a beginner question, what does levels do in the forecast function

fcsts_df = fcst.forecast(h=12, fitted=True, level=(90, 10))

Also if there's something to better learn how autoarima works

👋 Hello Nixtla team, wondering if you've come across AS197 estimator for SARMA models from this paper: Melard, G. "A fast algorithm for the exact likelihood of autoregressive-moving average." (1984). Thank you for the excellent numba SARMA implementation 🤝, it is definitely a dramatic improvement compared to existing alternatives! However, I found that it's still not fast enough as a benchmark method for less "well-behaved" time series: in particular with high frequency commodities time-series with high (p, q) and seasonal orders. This AS197 estimator, however, seems to solve this issue and the only open-source implementation at the moment is in Gretl I've been studying the performance of

auto_arima

(less numba compile time) and found a massive variance in training times (between 3-60 seconds) across time series as opposed to <5-10 seconds with

auto_ets

. After doing a meta-analysis of the fit-times against the optimal orders, I found that the current implementation is still quite slow at higher orders and with seasonal orders. Nevertheless, while looking through

statsmodels

, I found an issue that suggests using Chandrasekhar recursions (https://github.com/statsmodels/statsmodels/issues/6812) to dramatically speed up ARMA training times (up to >2-4x speed-up for higher orders with conditional SS). I've attached a screenshot from the paper that demonstrates this. Note: AS154 is the current Kalman filter implementation used in

statsforecast

and

fable

.

Hi. Sorry for bothering. Quick question: I tried running auto_arima_prophet in the same setup we have used Prophet for our forecasting and we have seen that auto_arima_prophet has 7x the running time for fit and predict, that Prophet has. Is there something we are doing wrong ? MSE is still 10% better for ARIMA, but still, I find it weird.

Hi everyone, I joined this Nixtla community slack channel just now. I am working on hierarchical time series project. I am following this https://www.kaggle.com/code/konradb/ts-8-hierarchical-time-series/notebook to implement the hierarchical model.

I am using the following command to install the required libraries. pip3 install hierarchicalforecast statsforecast datasetsforecast

Facing some errors and issues. Please someone from the community who has faced these issues or any kind of errors can help me. It is blocking my workflow since past 4 days.

I tried to install hierarchicalforecast library using the following command. pip3 install hierarchicalforecast