Moving Average Strategy

[Linh Nguyen]

Hi everyone,

My purpose: Determine The Best Moving Average Crossover Trading Strategy.(Moving average is calculated on the closing price)

Buy and sell signal:

• A buy signal: when a shorter-term moving average crosses above a longer-term moving average.
• A sell signal: when a short moving average crosses below a long moving average.

Process:

1. If a buy signal appears on date t, we will buy a stock at the opening price on date t+1
2. After buying a stock, we will wait for the appearance of a sell signal (e.g. on date t+10) to sell the stock at the opening price on the following date (t+11).
3. The holding-period returns is calculated as (selling price - buying price)/buy price.
4. The new round begins. Wait for the buy signal and so on...

Constraints:

• After buying a stock, the stock needs 3 days to be credited to our account, then we can sell it. For example, we buy a stock on date t, we can't sell the stock on date t+1, t+2. We only can sell the stock from date t+3.

My problem: I don't know the code to finish this task.

About data:

* Example generated by -dataex-. To install: ssc install dataex
clear
input float date long firm double(open_price close_price)
22446 1  83.8  83.2
22447 1  83.2  83.1
22448 1  82.1  82.6
22449 1  83.1  83.8
22452 1  83.6  85.6
22453 1  86.1  85.4
22454 1  85.3  84.9
22455 1  84.3  84.7
22456 1  84.7  85.2
22459 1  86.1  85.7
22460 1  86.6  88.6
22461 1    89    88
22462 1  88.1  88.9
22463 1  89.3  91.9
22466 1    95  92.7
22467 1  92.5    87
22468 1    87  90.5
22469 1  90.4  89.8
22470 1  89.6  89.7
22473 1  88.8  86.1
22474 1  86.9    87
22475 1  87.6  85.2
22476 1  84.7  85.4
22477 1  85.9  86.8
22480 1  85.2  84.7
22481 1  84.8  86.8
22482 1  86.8  86.6
22483 1  86.2  89.4
22446 2 102.1 102.8
22447 2 102.5   104
22448 2   104 104.2
22449 2 104.5 108.5
22452 2   108 106.6
22453 2 107.1 107.6
22454 2   108 109.7
22455 2   110 110.7
22456 2 110.7 112.6
22459 2 112.6 112.6
22460 2   112   114
22461 2   114 116.4
22462 2 116.7 116.4
22463 2 116.4 114.9
22466 2 113.9 113.6
22467 2 112.5   112
22468 2   112 113.4
22469 2 111.2   111
22470 2   110 109.8
22473 2 109.5 107.1
22474 2 106.9 103.4
22475 2 103.4 101.8
22476 2 100.2 102.5
22477 2 102.5 102.5
22480 2   100    97
22481 2  97.8 100.2
22482 2  99.5    99
22483 2  98.9 100.3
end
format %td date
label values firm firm
label def firm 1 "FPT", modify
label def firm 2 "VCB", modify

My code: I know how to use -tsegen- to calculate moving average:

tsegen ma2=rowmean(L(0/1).close_price) // moving average for 2 days
tsegen ma3=rowmean(L(0/2).close_price) // moving average for 3 days

However, I don't know how to use -forevalue- to do it, and how to calculate the holding-period returns and chose the highest one.

I hope that you could help me. Thanks in advance.

Best

[Linh Nguyen]

Hi everyone,

I find a way how to calculate the moving average:

* Calculate moving average all period
forvalues i = 3(1)10{
          tsegen ma`i' = rowmean(L(1/`i').close_price,`i')
}

* Create buy and sell signals
forvalues i = 3(1)10{
          forvalues j = 6(1)10{
           gen signal`i'`j' = 0
          replace signal`i'`j' = 1 if ma`i' > ma`j' & `i' +3 < `j'
          }
}

forvalues i = 3(1)10{
          forvalues j = 6(1)10{
           gen position`i'`j' = signal`i'`j' - signal`i'`j'[_n-1]
          }
}
/* position = 1, it implies that the Signal has changed from 0 to 1 => triggering a buy call
   Position = -1, triggering a sell call*/

However, I don't know how to calculate the holding-period return for the period from position = 1 to position -1 (the holding period).

Could anyone here help me?

Thanks in advance.

[Linh Nguyen]

Could anyone here help me?