Forecasting indicators - page 35

[Sergey Golubev]

Universal Regression Model for Market Price Prediction

The market price is formed out of a stable balance between demand and supply which, in turn, depend on a variety of economic, political and psychological factors that are difficult to be directly considered due to differences in nature as well as causes of their influence.

----------------

Universal regression model for market price prediction (Part 2): Natural, technological and social transient functions

As it turns out, the transient functions /1/ meant to analyze the material balance of the trading process are able to adequately describe other processes, both dynamic and static, in all areas of human life. The theory is based on three functions with three parameters. One of them (current C function) is differential, while two others are integral forms of the Gamma function distribution density. The past function P is obtained from C by integrating it from 0 to infinity with the n+1 parameter, while the future function F is used with the n parameter.

[Sergey Golubev]

Forecasting with ARIMA models in MQL5

The article Implementing an ARIMA training algorithm in MQL5, describes the CArima class for building ARIMA models. Although it is technically possible to use the class as it is to apply a model and make predictions, it is not intuitive. In this article we will address this shortcomming and extend the class to enable easier to use methods for applying models to make predictions. We will discuss some of the complications related to implementing predictions as well as some new features added to the class. To conclude we will use the complete class to build a model , use it to predict forex prices by applying it to an expert advisor and indicator.

[Sergey Golubev]

Forum on trading, automated trading systems and testing trading strategies

Trading Systems Based on Signal Indicators

Sergey Golubev, 2023.06.30 08:00

Holt's linear trend method

Holt (1957) extended simple exponential smoothing to allow forecasting of data with a trend. This method involves a forecast equation and two smoothing equations (one for the level and one for the trend):

1. Holt double exponential smoothing 2.2 indicator for MT4 (post #14301 to download)

Forum on trading, automated trading systems and testing trading strategies

Elite indicators :)

Mladen Rakic, 2016.12.01 10:01

Since Holt - Winters double exponential smoothing is used for explicit forecasting, this version is doing the forecasting part too.

The method used is the following

As with any forecasting method, be advised that the forecasting part is a subject of changes (it will recalculate/repaint) and use it accordingly (by all means, no signals or alerts on that part)

2. Holt double exponential smoothing trend indicator for MT4 (post #14306 to download)

Forum on trading, automated trading systems and testing trading strategies

Elite indicators :)

Mladen Rakic, 2016.12.01 17:10

Also the "trend" version of Holt's double exponential smoothing for metatrader 4 too

---------------

An I attached those two forecasting indicators to the charts:

[Sergey Golubev]

Reimagining Classic Strategies: Forecasting Higher Highs And Lower Lows

In this article, we closely examine the classic price action trading strategy of trading based on “higher highs” or “lower lows.” We trained various models to predict two distinct targets. The first target involved predicting changes in price, the simplest target possible. The second target aimed to determine whether the future close price would be higher than the current high, lower than the current low, or between the two.