# AR Extrapolation Of Price Indicator For MT5 Please note: This strategy was publicly published in the trading community and is free to use. We do NOT make an attempt to decide if this strategy is profitable or not, because we know that the major factors regarding trading results are the skills/experience of the trader who executes the strategy. Therefore, we are mainly explaining the components and rules of the strategy. If applicable, we are highlighting advantages, disadvantages and possible improvements of the strategy.

## AR Extrapolation of Price Introduction

The AR extrapolation of price indicator for MT5 is derived using the autoregressive model. The term “AR” in this indicator stands for auto-regression, meaning future values are derived from past values. The process used to obtain these future values is known to mathematicians and statistical analysts as linear regression. At some point in time, this method of analyzing data seems to have found its way into the world of custom indicators.

The above visual shows the AR indicator with the default settings. The line in blue and red is the AR extrapolation of price indicator as it is applied on GBPJPY four-hour chart with the chart shift fully extended to the left. By default, the red line can extend to the right up to 150 candles into the future. The user can adjust this value in inputs and may opt to display a small portion of the red line since what is important in trading is the potential direction of price in the near future. ## FREE AR extrapolation of price Indicator

To receive my email 100% sure:
Put my email on your whitelist!

### What Is AR Model?

To use this indicator in trading, one must get an idea of the AR model, though this knowledge is not a requirement. Basically, an AR model is a method used to generate current or future data in a time series. This is done using past data in the same time series. In this context, a time series means the sequence or arrangement of data in a graph, such as movement of price on the chart of a financial security, over a period of time.

In AR model, we use past data in a time series to model and predict future behavior provided that there is correlation between currently selected values and those preceding or succeeding them. This method works due to the assumption in technical analysis that the past has some influence over or affects the potential direction of price in the future. In linear regression, there is assumption that natural phenomena have linear relationship, meaning the behavior can be represented by a line. However, as this indicator shows, this line does not need to be a straight line. It can be a curved line as the AR extrapolator does. ### AR Extrapolation of Price Indicator Parameters

The AR indicator that is provided in this website uses four parameters that the user can adjust at will in inputs either at the time of launching the indicator or at any time he prefers. These parameters with their default values are presented and explained below:

• UseDiff (Use price difference) = false – By default, this variable is set to false, which means that actual prices are used, not the price differentials. The above graphic shows how the indicator looks like if the UseDiff variable is set to true, meaning price differences are used.
• Ncoef (Model coefficients) = 150 – This number directly affects how the blue and red lines display on the chart. Setting a higher value would reduce the amplitude of the line; a lower value would increase the amplitude. The user can explore which value is better (i.e., higher, lower or default).
• Nfut (Future bars) = 100 – This value means that the red line will be drawn for 100 bars into the future starting from the current candle. The length of the red line shown depends on the chart shift though. The image below uses a value of 20 for this variable.
• kPast (Past bars in increments of Ncoef) = 3 – This is the number of bars in the past being used for prediction of future values. ### AR Extrapolation of Price Components

The indicator is plotted on the main chart and directly interacts with price. It is just one line that changes color depending on its location.

• Blue line – represents the historical and recent prices of the asset
• Red line – represents the extrapolated prices of the asset in the future ## AR Extrapolator versus Fourier Extrapolator

The Fourier extrapolator makes use of an algorithm to obtain harmonic frequencies. What it does is it adds the harmonic sequence of trigonometric series until a predefined total number of harmonics is achieved. As can be seen in the image above, the Fourier extrapolator likewise uses two colors to present historical and future values. The aqua line represents modeled past values, while the gold line represents modeled future values.

As with the AR extrapolator, the Fourier extrapolator uses almost the same set of settings. The two types of price extrapolation indicators present differently on the chart. While the AR extrapolator appears jagged, the Fourier extrapolator looks smooth. As to which indicator is better, this is an individual judgement call. Normally, both indicators can be used the same way in trading as explained in the following section. ## How to Trade with the AR Extrapolation of Price

Since the AR Extrapolation of Price indicator attempts to predict what will happen in the future, the best way to use this technical tool is simply to follow the line. Here we are talking about the red line, which models future prices. A simple trading strategy is presented below:

• Buy entry – Open a buy position when the red line is going up from the position of the current open candle.
• Sell entry – Open a sell position when the red line is going down from the position of the current bar.
• Trade exit – Closing trades can be a personal preference. The suggested method is to set the take profit target based on the trade risk. Taking profit at a point equal to two times the risk is a good bet. Still, it requires a lot of testing to be able to determine the best exit strategy.

Using the AR extrapolation indicator alone as a trading tool is not recommended as the tool is new in technical analysis. Further testing of the indicator via demo trading can give one an insight into the usefulness of this technical tool.

## A Closer Look at the AR Model

The AR model is implemented based on the assumption that the current sequence of events has a direct relationship or is a result of previous or past events. It employs past results to predict future values in a time series.

If one event is said to be affected in direct proportion with another event or if one event results in the occurrence of another event with some level of certainty, then the two events are said to be positively correlated.

One area in which positive correlation is applicable is forecasting asset prices, that is, when you want to find out how historical prices of an asset can affect future prices. Another area is the movement of correlated pairs in the foreign exchange market. One example is AUDUSD and NZDUSD. While the two instruments do not move with the same strength or volatility, they seem to move together most of the time.

### Prediction Methods of AR Models

As listed below, AR models use three prediction methods.

1. Time series prediction method
2. Qualitative prediction method
3. Casual model prediction method

In this article, only one method is discussed due to space limitation. The time series prediction method utilizes time as an independent factor to generate demand. The prediction is done using patterns of historical data only. Past data is being used to create a model of time series that is utilized to judge future values. In this case, calculations are made over successive periods or successive points. These calculations or measurements can be taken on specific intervals such as hourly, daily, weekly, monthly, and even yearly.

### Benefits of AR Models

There are many benefits of using AR models, namely:

• The correlation feature can be used to find out if there is randomness or lack thereof.
• The AR model can predict repetitive patterns in data.
• Less data is needed to predict the outcomes. 