Thank you for sharing your work. I notice the GMC projection for the DJIA varies between this post and the one you included on 8 Oct, yet both are labelled forecast 3. Is one mislabeled?
Thank you Andrew. The forecast labelled with forecast 3 is just the same. Forecast 3 is the model I use to do this forecasting. I developed this 13 years ago. The differences are the dynamic updates ones the actuals come in. The projected forecast is based on past cycle behaviour that will change from time to time, due to the actual market behaviour and past amplitude of this cycle. Both projections were made by the same Forecast model 3.
Got it, cheers. So the labelled pivots are important based on longer term cycles and may suggest H or L. The forecast curve is updated on a daily basis based on the days price action but is not near term predictive in advance as polarity will change from time to time, is that correct?
Hi Andrew, the labels were only a few lines in the sand where I expected a change in trend back in early July. I did not change them from early July. Yes this cycle can invert, extend or contract from time to time but not that often. The dynamic update and forecast algorithm is not as you assumed and remains a secret. I have never measured whether near term prediction is more accurate than prediction on a longer term and my observation is it varies from time to time. Should be an interesting study to do. I did have analysed whether there is a bias in the forecast evaluating putting it through the Mincer-Zarnowitz regression test as a good forecast should not have a bias. To the surprise of the academic researcher it past the test positively. There seem to be no bias in the forecast over the time measured from 2008-2020. It does not mean the forecast will always be correct though. I use the chart as a seasonal indicator and I let the actual chart in front of me decide (not the forecast).
I would like to see gold, bitcoin and oil. Thanks.
Hi John, will do. It is on my list.
Major indices $SPX / $DJI
Thank you for sharing your work. I notice the GMC projection for the DJIA varies between this post and the one you included on 8 Oct, yet both are labelled forecast 3. Is one mislabeled?
Thank you Andrew. The forecast labelled with forecast 3 is just the same. Forecast 3 is the model I use to do this forecasting. I developed this 13 years ago. The differences are the dynamic updates ones the actuals come in. The projected forecast is based on past cycle behaviour that will change from time to time, due to the actual market behaviour and past amplitude of this cycle. Both projections were made by the same Forecast model 3.
Got it, cheers. So the labelled pivots are important based on longer term cycles and may suggest H or L. The forecast curve is updated on a daily basis based on the days price action but is not near term predictive in advance as polarity will change from time to time, is that correct?
Hi Andrew, the labels were only a few lines in the sand where I expected a change in trend back in early July. I did not change them from early July. Yes this cycle can invert, extend or contract from time to time but not that often. The dynamic update and forecast algorithm is not as you assumed and remains a secret. I have never measured whether near term prediction is more accurate than prediction on a longer term and my observation is it varies from time to time. Should be an interesting study to do. I did have analysed whether there is a bias in the forecast evaluating putting it through the Mincer-Zarnowitz regression test as a good forecast should not have a bias. To the surprise of the academic researcher it past the test positively. There seem to be no bias in the forecast over the time measured from 2008-2020. It does not mean the forecast will always be correct though. I use the chart as a seasonal indicator and I let the actual chart in front of me decide (not the forecast).
Kindly include NQ and SPX as well. Thank you!
Hi Melwyn, they will be included as well.