Regression Algorithm while passing in future values?
Regression Algorithm while passing in future values?
Loading saved threads...
Ryan Gaudion · External communityPost link
External question — Data Science Stack Exchange
Author: Ryan Gaudion
Original post: https://datascience.stackexchange.com/questions/115602
License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/
Adaptation: HTML converted to plain text; contact email addresses removed.
Very similar to this question here:
https://stats.stackexchange.com/questions/406416/including-future-values-in-a-regression
Is it possible to pass future (expected) values into a regression algorithm in order to tune the results.
For example, training & predicting stock market prices but with 3 predictions, 1 if macroeconomic factor increases by x% (such as inflation or interest rates), 1 if it decreases by x% and 1 if it stays the same?
What regressions machine learning algorithms allow us to pass in the future expected values to see multiple "what-if" predictions?
EDIT: To explain a bit further. I would like to predict Time-Series based data while passing in an additional independent variables (other than time). Is this possible & if so, which algorithms allow it?
Quote
Report
Nicolas Martin · External communityPost link
External answer — Data Science Stack Exchange
Author: Nicolas Martin
Original post: https://datascience.stackexchange.com/a/115643
License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/
Adaptation: HTML converted to plain text; contact email addresses removed.
Is it a multi-variate time-series forecast?
If yes, you can add several features that could improve predictions.
The most common one is using LSTM neural networks:
https://machinelearningmastery.com/multivariate-time-series-forecasting-lstms-keras/
But you might have better results with XGBoost:
https://cprosenjit.medium.com/multivariate-time-series-forecasting-using-xgboost-1728762a9eeb
If you have several features that don't change much with time, random forest could be interesting:
https://towardsdatascience.com/multivariate-time-series-forecasting-using-random-forest-2372f3ecbad1
In every case, some data analysis to know the correlations in your data and some patterns could be necessary before choosing the right algorithm:
https://www.analyticsvidhya.com/blog/2021/10/a-comprehensive-guide-to-time-series-analysis/
Quote
Report
Post Reply
Quoted from Forex.com.bd-Editorial External question — Data Science Stack Exchange Author: Ryan Gaudion Source score (net votes, not local likes): 0 Original post: https://datascience.stackexchange.com/questions/115602 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. Very similar to this question here: https://stats.stackexchange.com/questions/406416/including-future-values-in-a-regression Is it possible to pass future (expected) values into a regression algorithm in order to tune the results. For example, training & predicting stock market prices but with 3 predictions, 1 if macroeconomic factor increases by x% (such as inflation or interest rates), 1 if it decreases by x% and 1 if it stays the same? What regressions machine learning algorithms allow us to pass in the future expected values to see multiple "what-if" predictions? EDIT: To explain a bit further. I would like to predict Time-Series based data while passing in an additional independent variables (other than time). Is this possible & if so, which algorithms allow it?
Checking account access…