<![CDATA[Stories by Sarem Seitz on Medium]]> https://medium.com/@sarem-seitz?source=rss-8f6d033b1a40------2 https://cdn-images-1.medium.com/fit/c/150/150/1*0Bb8Cw1jMDkSO65hXzD6pw.png Stories by Sarem Seitz on Medium https://medium.com/@sarem-seitz?source=rss-8f6d033b1a40------2 Medium Sat, 14 Nov 2026 02:58:00 GMT <![CDATA[Winning with Simple, not even Linear Time-Series Models]]>

If your dataset is small, the subsequent ideas might be useful

Continue reading on TDS Archive »

]]> https://medium.com/data-science/winning-with-simple-not-even-linear-time-series-models-6ece77be22bc?source=rss-8f6d033b1a40------2 https://medium.com/p/6ece77be22bc Wed, 10 May 2023 06:38:13 GMT 2023-05-11T13:46:46.647Z <![CDATA[Varying Coefficient GARCH]]>

Let’s make GARCH have varying coefficients to handle non-linear conditional variance.

Continue reading on DataDrivenInvestor »

]]> https://medium.datadriveninvestor.com/varying-coefficient-garch-7880c8105b01?source=rss-8f6d033b1a40------2 https://medium.com/p/7880c8105b01 Thu, 19 Jan 2023 22:04:08 GMT 2023-01-22T01:59:29.288Z <![CDATA[When Point Forecasts Are Completely Useless]]>

While point forecasts are very popular, be aware of some unlucky pitfalls

Continue reading on TDS Archive »

]]> https://medium.com/data-science/when-point-forecasts-are-completely-useless-79cd27d0b1e5?source=rss-8f6d033b1a40------2 https://medium.com/p/79cd27d0b1e5 Sun, 01 Jan 2023 22:33:06 GMT 2023-01-04T14:22:01.537Z <![CDATA[Why I prefer probabilistic forecasts — hitting time probabilities]]>

Point forecasts are good for making decisions. With probabilistic forecasts, you can also make the right ones.

Continue reading on TDS Archive »

]]> https://medium.com/data-science/why-i-prefer-probabilistic-forecasts-hitting-time-probabilities-726667b9c91d?source=rss-8f6d033b1a40------2 https://medium.com/p/726667b9c91d Tue, 06 Dec 2022 22:17:55 GMT 2022-12-09T15:07:01.418Z <![CDATA[Random Forests and Boosting for ARCH-like volatility forecasts]]>

Tree models are not just useful for point and mean forecasts.

Continue reading on DataDrivenInvestor »

]]> https://medium.datadriveninvestor.com/random-forests-and-boosting-for-arch-like-volatility-forecasts-21b601442494?source=rss-8f6d033b1a40------2 https://medium.com/p/21b601442494 Fri, 07 Oct 2022 05:42:31 GMT 2022-10-08T00:56:54.164Z <![CDATA[Forecasting with Decision Trees and Random Forests]]>

Random Forests are flexible and powerful when it comes to tabular data. Do they also work for time-series forecasting? Let’s find out.

Continue reading on TDS Archive »

]]> https://medium.com/data-science/forecasting-with-decision-trees-and-random-forests-ac081ff451ec?source=rss-8f6d033b1a40------2 https://medium.com/p/ac081ff451ec Mon, 19 Sep 2022 10:01:41 GMT 2022-09-21T18:50:51.863Z <![CDATA[Multivariate GARCH with Python and Tensorflow]]>

One primary limitation of GARCH is the restriction to a single dimensional time-series. In reality, however, we are typically dealing with…

Continue reading on DataDrivenInvestor »

]]> https://medium.datadriveninvestor.com/multivariate-garch-with-python-and-tensorflow-8fb746c3fec9?source=rss-8f6d033b1a40------2 https://medium.com/p/8fb746c3fec9 Sun, 11 Sep 2022 20:55:34 GMT 2022-09-12T06:53:40.865Z <![CDATA[Cointegrated time-series and when differencing might be bad]]>

You have heard about integrated time-series data but what about cointegration?

Continue reading on TDS Archive »

]]> https://medium.com/data-science/cointegrated-time-series-and-when-differencing-might-be-bad-f4e2a9132a0c?source=rss-8f6d033b1a40------2 https://medium.com/p/f4e2a9132a0c Thu, 25 Aug 2022 09:13:55 GMT 2022-08-25T13:57:00.107Z <![CDATA[Facebook Prophet, Covid and why I don’t trust the Prophet]]>

Facebook Prophet is highly popular for time-series forecasting. Let me show you why I am not a big fan and what else you can use.

Continue reading on TDS Archive »

]]> https://medium.com/data-science/facebook-prophet-covid-and-why-i-dont-trust-the-prophet-f70d367fb13a?source=rss-8f6d033b1a40------2 https://medium.com/p/f70d367fb13a Mon, 08 Aug 2022 10:12:14 GMT 2022-08-09T13:23:47.337Z <![CDATA[Probabilistic CUSUM for change point detection]]>

CUSUM is arguably the simplest algorithm for change point detection problems as in IoT or finance applications.

Continue reading on TDS Archive »

]]> https://medium.com/data-science/probabilistic-cusum-for-change-point-detection-121f793ab3a1?source=rss-8f6d033b1a40------2 https://medium.com/p/121f793ab3a1 Thu, 04 Aug 2022 13:09:43 GMT 2022-08-04T15:10:30.812Z