CHAPTER 4: CREATING TRADING STRATEGIES

To create a trading strategy, you need to identify a set of rules that will determine when to enter and exit trades. In this chapter, we will cover some basic trading strategies and show you how to implement them in Excel using formulas and VBA.

For example, have you ever heard of the “Three White Soldiers” and “Three Black Crows” stock candlestick patterns? Let’s take a look at this chart:

This is the so-call "Three White Soldiers". It is a bullish candlestick pattern that typically indicates a strong uptrend in the stock price. It consists of three long white (or green) candles, each with small or no shadows, that open higher and close higher than the previous day's candle.

The pattern is called "Three White Soldiers" because the three candles resemble three soldiers marching in a row. It is a sign of strong buying pressure and suggests that the bulls are in control of the market. Traders often look for this pattern to identify potential entry points for long positions, but it should be confirmed with other technical indicators before making any trading decisions.

Now, here we have another pattern:

Let’s say hello to the "Three Black Crows". It is a bearish candlestick pattern that is typically seen at the end of an uptrend in the stock price. It consists of three long black (or red) candles, each with small or no shadows, that open lower and close lower than the previous day's candle.

The pattern is called "Three Black Crows" because the three candles resemble three crows sitting in a row. It is a sign of strong selling pressure and suggests that the bears are in control of the market. Traders often look for this pattern to identify potential entry points for short positions, but it should be confirmed with other technical indicators before making any trading decisions.

According to these ideas, we can develop a trading strategy to buy stocks while we see the Three White Soldiers appearing, and short sell the stocks when we find the Three Black Crows sitting there.

Now we start to write a VBA Macro, a simple program to execute this strategy. To write a VBA macro, follow these steps:

1.       Open Microsoft Excel.

2.       Press Alt+F11 to open the Visual Basic Editor.

3.       In the Visual Basic Editor, select the workbook in which you want to create the Macro.

4.       Click on the Insert menu and select Module. This will create a new module for the Macro.

5.       In the module, type the code for the Macro. The code should be written in the VBA programming language.

6.       Save the Macro by clicking on the Save button or by pressing Ctrl+S.

7.       Close the Visual Basic Editor.

8.       To run the Macro, open the workbook in which it was created and press Alt+F8 to open the Macros dialog box.

9.       Select the Macro you want to run and click the Run button.

Note: It is important to test the Macro thoroughly before using it on important data. Also, make sure to enable Macros in Excel before running the Macro.

Before starting to do this VBA coding, we need a plan. It looks like the programming flowchart below:

First of all, we need to declare variables. Declaring variables in VBA is important because it helps to allocate memory to store the data that the variable will hold. This allows the program to use the variable efficiently and effectively.

By declaring variables, you can also specify the data type of the variable (such as integer, string, or boolean), which ensures that the variable is used correctly throughout the program. If you don't declare a variable, VBA will assume a data type based on the value assigned to the variable, which can cause errors or unexpected results.

Declaring variables also makes the code more readable and easier to understand, as it makes it clear what each variable represents and how it will be used.

 In the Main Procedure, we will turn off Automatic Calculation to speed up VBA. Use  Application.Calculation = xlCalculationManual at the beginning of the code, and turn it on again at the end with Application.Calculation = xlCalculationAutomatic.

Then we do some initialization, and call four subroutines, each implementing a different function.

In the Self-Verifying procedure, we will verify all formulas used in every Excel worksheet, such as calculating moving average value, indicator value, stock position, cash, equity, etc. These formulas have been set into the demo Excel files. To prevent them from being ruined or damaged accidentally, every time the VBA Macro is run, the program will verify all formulas automatically to ensure that the results and output are correct and fit the same standards.

In the Digging procedure, we will dig through every parameter combination to simulate opening and closing positions, calculate the profit or loss, compare ROI, eliminate low-performance parameters, and select profitable combinations.

For digging out good parameter combinations, we want to borrow some ideas from Machine Learning technology (very fashionable, eh?), to divide historical data into two parts, test our Trading Strategy (or we can use another word, “Model”, in Machine Learning) under different conditions.

In machine learning, the data is often divided into two subsets: the training set and the testing set. The training set is used to train the machine learning model, while the testing set is used to evaluate the performance of the trained model.

The purpose of dividing the data into training and testing sets is to avoid overfitting, which is a situation where the model performs very well on the training data but poorly on new, unseen data. By evaluating the model's performance on the testing set, we can get an estimate of how well the model will perform on new, unseen data. For stock trading, it means: based on the training set’s performance, we can select some strong parameters to test in the testing set, screen out and throw out the weak ones which will lead us to failure and loss, those survivors will give us a likelihood to estimate how profitable the strategy will perform in the future.

The historical data can be divided into training and testing sets, with a typical split of 70% for training and 30% for testing. In our Demo Excel files, the default split is 70% for Training set and 30% for Testing set. Users can adjust them to 60% or 80% for training and 40% or 20% for testing. Be careful, the split ratio could affect the final results significantly.

The most important part in Digging procedure is the trading strategy conditions. We have a subroutine named as Signals to take charge of comparing the opening conditions. For example, we can Call Signals() to generate the simplest "Three White Soldiers and Three Black Crows" signals (three white candles or three black candles in a row) like this:

VBA Macro will execute these codes again and again, until finish all loops which are appointed by us. According to all results, the program will delete bad parameters and retain the good ones to pass to the next procedure.

Scanning procedure looks as same as Digging procedure. All good parameters screened out from the Digging procedure will be tourtured again in this part. We will use histrorical data in Testing set to test parameters, simulate opening and closing positions during testing set’s time period. Most parameter combinations will be elimated because failing to get profit or the the profit too low. Only those top performance players have value to be thought as a real fighter. At the final, we will let the winners scanned in Testing set run a full test, since day one to the most rencent data. And screen out the final competitor, for a future’s real battle. 

The Scanning procedure looks the same as the Digging procedure. All good parameters screened out from the Digging procedure will be tortured (if you feel this word is too cruel, replace it by “tested”) again in this part. We will use historical data from the Testing set to test the parameters and simulate opening and closing positions during the Testing set's time period. Most parameter combinations will be eliminated because they fail to generate a profit, or the profit is too low. Only the top-performing parameters have value and can be considered as real fighters. In the end, we will let the winners scanned in the Testing set run a full test, from day one to the most recent data, and screen out the final competitors for future's real battles.

In the final stage, the VBA Macro generates a chart which allows users to easily evaluate the results. With just a glance at the curve, the pattern speaks for itself, better than thousands of words. The curve may resemble something like this:

Can you guess which stock this is? If I told you it was Tesla, you might say "No wonder", eh? The curve appears like a roller coaster ride, eh? The ROI is so high (Annul ROI goes to almost 40%), but it will also cause our blood pressure to rise high, eh? The “Three White Soldiers” and “Three Black Crows” trading strategy looks “WOW!”, but still require a lot of modifications and optimizations, which we will cover in a future chapter.

As we come to the end of this chapter, let me remind you of a crucial fact: trading is not for everyone. If you find yourself getting anxious or losing sleep over your trades, it's time to take a break and re-evaluate your life choices. Maybe knitting or stamp collecting is more your speed.

But if you're still determined to become a successful trader, remember to keep your sense of humor intact. As Groucho Marx once said, "If you're not having fun, you're doing something wrong." So laugh off your losses, celebrate your wins, and don't forget to take a break from staring at those candlestick charts every once in a while.

With that said, happy trading and may the Three White Soldiers always march in your favor! (Just don't get too close to the Three Black Crows.)