# Seasonality of the US Utilities Sector

04 July 2022We continue to study seasonal patterns of changes in the value of financial instruments by sectors of the US stock market.

And today we will pay attention to the sector of utilities.

We will identify the seasonal trends of the S&P 500 Utilities Index and examine the trading signal based on its seasonal patterns.

The utilities sector index - S&P 500 Utilities - includes 28 companies, the major representatives of which are:

- Nextera Energy
- Duke Energy
- Southern Company
- Dominion Energy
- Exelon

Changes in the market value of the US utilities sector index (S&P 500 Utilities) have pronounced seasonal manifestations. And a trading signal based on the seasonal patterns of the index is profitable.

Historical data of S&P 500 Utilities index quotes:

- Timeframe - МN (month);
- Period: October 1989 - September 2021;

There are 384 values in total.

**To identify the presence of seasonality, we will use the following criteria:**

1. The share of cases of positive or negative monthly changes for the period under review is more than 53%.

2. The ratio of the average value of positive monthly changes to the average value of negative monthly changes:

- greater than 1, when the share of positive monthly changes is greater than 53% (see point 1);
- less than 1, when the share of negative monthly changes is greater than 53% (see point 1).

3. Linear trend in the distribution of monthly changes over the past 10 years:

- upward trend, when the share of positive monthly changes is greater than 53% (see point 1);
- downtrend, when the share of negative monthly changes is greater than 53% (see point 1).

If all three conditions are met, we will talk about the presence of seasonality in the price changes of a financial instrument.

The section of historical data from 1989 to 2010 will be the base period on which we will identify seasonal patterns in value changes. If such patterns are identified, they will be used as signals to buy or sell in the tested section of history from 2011 to 2021. Thus, we will check the effectiveness of the identified trading signals based on seasonal patterns.

** **

**Analysis of the obtained results **

** **

**Seasonal changes in the S&P 500 Utilities index (price changes per month, in %)**

*Explanation for the final lines:*

*1) In the first line, the share of cases of positive changes was calculated based on the analysis of monthly price changes. On the example of January, it looks like this: we take all the months "January" for 22 years, so we get 21. Of these 21 months, a price growth has been noted on the basis of 9 months. That is, the share of growth cases has been 43%.*

*2) In the second line, we compare the average value of all positive monthly changes with the average value of negative changes.*

*If the value obtained is greater than 1, then the price passes more points in the growth months than in the months of declines.**If the value obtained is less than 1, then in the months of decline the price passes more points than in the months of growth. In the case of the month "January", the value is less than 1.*

So, seven months (January, February, March, April, May, November and December) meet two of the three criteria for pronounced seasonality.

Let's check these months according to the third criterion - the direction of the linear trend of the distribution of monthly changes for 2000-2010:

So, the linear trend of changes in January over 10 years is directed downwards, which satisfies the last criterion for identifying seasonality.

The 10-year trend of February, March, April and November is directed downwards, which refutes the assumption of a seasonal uptrend in prices.

The linear trend of May and December changes over 10 years is directed upwards, which does not satisfy the last criterion for identifying seasonality.

Thus, with some confidence, we can say that the index of the utilities sector tends to decrease in January.

These seasonal patterns we will use as a trading signal for testing on historical data sampling from 2011 to 2021.

We will evaluate the trading signal according to the following criteria:

**The****rate of return**reflects the relative change in the quotes of financial instruments in percentage. A positive value of the rate of return indicates the profitability of the strategy, negative - about the loss.

The rate of return (R) of a financial instrument is calculated using the formula:

**R = Σ P (%) / n,**

where:

n is the number of transactions;

P (%) – the percentage of change in the quote of a financial instrument at the time of fixing a position, is calculated as follows:

for buy positions

P (%) = (position closing price - position opening price) / position opening price * 100%

for sell positions

P (%) = (position opening price - position closing price) / position opening price * 100%

- The
**average rate of return of profitable transactions (AR)**includes the rate of return of only profitable transactions, as a percentage:

**AR = Σ D (+) / n,**

where:

n is the number of profitable transactions;

D (+) – rate of return of profitable transactions.

**Average drawdown (AD)**reflects the average loss when closing losing transactions for the entire trading period, as a percentage. The lower the value of the average drawdown, the lower the losses, and the better the trading signal works.

**AD = | Σ D (-) / n |**

where:

n is the number of losing transactions;

D (-) – rate of return of losing transactions.

**Maximum rate of return (MaxR)**is the maximum profit from closing successful transactions for the entire trading period, as a percentage. The higher the maximum rate of return value, the better the trading signal works.

**MaxR = max (R)**

**Max drawdown (MD)**is the maximum of losses when closing unsuccessful transactions for the entire trading period, in percent (minimum profitability). The lower the value of the maximum drawdown, the better the trading signal works.

**MD = | min (R) |**

**Share of profitable positions (SPP)**shows the share of profitable trading positions from the total number of positions, as a percentage. The higher the DPP, the more profitable trades are made.

**SPP = number of profitable positions / total number of positions * 100**

The results of the strategy of selling the index of the utilities sector (S&P 500 Utilities) in early January and buying it at the end of the month are presented in the diagrams:

So, the rate of return of buying the S&P 500 Utilities index at the beginning of January and selling it at the end of the month was -1.8% with an average rate of return of profitable transactions and a drawdown of 2.6% and 3.4%, maximum rate of return and drawdown of 3.7% and 6 .5% respectively.

The share of profitable positions in March amounted to 27.3%.

The S&P 500 Utilities index tends to decrease in January.

Changes in the market value of the US Utilities Index are subject to seasonal fluctuations.

**The effectiveness of trading signals based on the seasonal patterns of the S&P 500 Utilities index has not been revealed.**

Detailed results are shown in the Appendix:

*Seasonality of the US Utilities sector.xlsx*

See also:

Comments

4