Predictive Model Accuracy

Discussions center on the use, reliability, and skepticism surrounding machine learning and statistical models for forecasting future events based on historical data, including time series predictions and methods like regression or neural networks.

📉 Falling 0.4x AI & Machine Learning
3,532
Comments
20
Years Active
5
Top Authors
#8950
Topic ID

Activity Over Time

2007
5
2008
21
2009
41
2010
78
2011
81
2012
70
2013
88
2014
127
2015
118
2016
179
2017
227
2018
252
2019
260
2020
309
2021
312
2022
285
2023
398
2024
381
2025
278
2026
22

Keywords

AI NN US GBDT ETS GAM ML VIX predictious.com LTV prediction predictions regression time series predicting predict model predictor models judgement

Sample Comments

akoboldfrying Jan 15, 2025 View on HN

Don't we need models to make predictions about the future?

WithinReason Apr 12, 2022 View on HN

You can't make accurate predictions without some kind of model

linkingday Feb 26, 2019 View on HN

What prediction algorithm are you using for this? Not sure if it's mentioned elsewhere on the site

petra Oct 8, 2016 View on HN

Probably feed it with historical data and see if it can tell what has happened.

rhizome Dec 18, 2016 View on HN

Sounds like a "prediction" just like the results you get from training your stock picker on historical data.

itry Oct 3, 2012 View on HN

A predictor doesnt have to be always right to be a predictor.

dia80 Jun 2, 2023 View on HN

Forecast! Time to reach for your linear regression, GBDT or NN.

hornd Apr 9, 2012 View on HN

This is fantastic, thank you. Any chance of adding predictions based on past performance, etc?

QuercusMax Jul 9, 2025 View on HN

Is there a reason you believe the models can accurately predict this sort of thing?

AJ007 Aug 8, 2019 View on HN

* this is assuming the past continues to be a predictor of the future