A decision is made to determine if one asset is undervalued or overvalued is relative to the other, based on small time interval key factor:

\ mathrm {Deviation} _ {(A / B)} = \ mathrm {Ratio} _ {(A / B) _t} – \ mathrm {SMA} _ {(A / B) _t}

\ mathrm {SMA} _ {(A / B)} = \ frac {\ sum_ {i = 1 } ^ n \ mathrm {Ratio} _ {(A / B) _t}} {n}

where:

n – the number carried in the averaging period t .

If the current deviation of the ratio of its moving average depth is in the negative area, it means that the activity is located in the numerator is underestimated relative to the asset located in the denominator.

Conversely, if the current deviation of the ratio of its moving average depth is in the positive area, it means that the activity located in the numerator is overestimated relative to the asset located in the denominator.

Analyzing graphs and defining a pair of under-rated and over-valued asset, at the same time takes longer.

At the same time, the proportion of each asset in the portfolio should be calculated on the basis of both the Beta assets at the time of the transaction.

In addition, since the value of Beta is not constant, it is periodically necessary to adjust the structure of the portfolio, to maintain it in a market- neutral state . When the ratio of assets to reach its average value is reverse transactions with a view to profit .

Despite the seeming simplicity of the strategy, it has its own pitfalls. The fact that the absolute deviation does not take into account the volatility of relations that represents certain risks.

For example, the absolute deviation may show an opportune moment to enter a position, as it is at the maximum or minimum values of the historical forming support and resistance levels.

But if you currently have ratios reflecting increased volatility, it is likely that these levels can be punctured. Considering the risk of increased volatility allows deviation ratio, expressed as a standard deviation.

During periods of relatively stable ratio between the value of assets , entering a position at a small value can be absolute deviation, because despite the lower returns from one transaction, the relationship often reverts to the mean, providing a greater number of possible transactions, thereby increasing profitability.

Conversely, during periods of instability ratio, the deviation becomes larger, and it is returned to the medium at least in connection with transactions than is recommended to make smaller and more distant from the average.

Another important factor is the relative strength and direction of the moving average . If its value is located a considerable distance from zero and coincides with the absolute deviation, it indicates possible collapse of the deviation.

But not at the expense of the return ratio to the average, and bringing it to the very middle of the current ratio. If the direction is opposite to the current moving average absolute deviation, it is very likely return ratio to its average value in the near future.

Trade in pairs in the deviations is most effective for short-term time intervals (within a few days) to more of the same long periods it is advisable to trade in pairs at the boundaries of the historic channel relationships formed by the historical minimum and maximum. Of the channel are the powerful support and resistance levels.

While the narrower the width of the channel in a pair, the greater the number of extremes to “test ” the strength of these boundaries, the more attractive the pair to trade. As a criterion when comparing the width of the historic canal in different pairs , used the percentage of historic highs to historic lows .