Here, Pandas package is used to deal with data, the NumPy package is used to work with arrays, the Requests package is used to pull data using DSS API, the Math package is used for mathematical functions, the SciPy package is used for complex functions, and few other packages are also used for support.

After importing all the required packages into our python environment, let's define a function to authenticate connection with DSS API.

At this step, we will take DSS username and password as input from user.

Extracting List of all FTSE 100 Stock’s

Next, we are going to extract the list of all FTSE 100 stock’s using 0#.FTSE ChainRic from DSS API. For this let's define a function to expand this ChainRic.

In the above code, we have used DSS API to extract the RICs and it's status of the stocks available in the FTSE 100 by sending request to DSS's RestAPI using request package. This function will post the request to HistoricalChainResolution and will return the data in dataframe format.

We first extracted the RICs and Status of the stocks in the FTSE 100 using the ExpandChain function and stored it into the ftse_ric dataframe. Then we defined a variable ‘ftse_list’ to store the extracted data into a NumPy array.

To learn more about "How to expand Chain RIC using the Tick History REST API in Python", visit here.

Pulling Intraday data of all the stocks in the FTSE using DataScope Select API

We have seen list of all the FTSE 100 stock’s RICs, now let’s pull some intraday data using DSS API.

For this let's define a function to extract data first for the given ChainRic, i.e., #0.FTSE.

In the above function as well, we have used DSS API to extract the data for all the RICs of FTSE 100 by sending request to DSS's RestAPI using request package. We first setup the request header and post the request to ExtractRaw to fetch data.

The fetched data will be written to a csv file using gzip compression, which will be loaded to a dataframe and returned.

In case, if there will be any exception it will be raised, for example if the request's returned status is not Accepted or Ok.

Here, we have defined retryInterval variable for re-requesting the status after it's value (in second) to check if post request is already completed.

Here, we can see the Status and progress of the ExtractRaw function. Once the data is fetched completely, few metadata for each of the RICs is also displayed such as #RIC, Domain, Start, End, Status, and Count. We can also view all the data values fetched using the last url printed at the end.

If you don’t know how to pull stock data using DSS API, I suggest to read article on it here.

Data Pre-Processing

At this step, we will pre-process the data to get it ready in the desired format.

1. First format the date-time column and select the columns which we need further for our strategy.

2. Transpose the dataframe so that RICs name become the column names, set the date-time column as index, and Close values in there corresponding columns.

Since, now the data is ready in the desired format, now we will save the extracted intraday data of all FTSE 100 stocks. We also dropped column with all null values. It is an optional step but a good practice to save time, so first saving the extracted intraday data in the name of ‘ftse100’ using the ‘to_csv’ function provided by the Pandas package.

Calculating Percentage Change and Momentum

Next we are the going to calculate change in percentage and momentum of all the stocks

First, we created an empty list named dc to store the percentage change of each stock. Using for loop, we are iterating over the RICs of all the stocks in the FTSE 100 and appended the calculated percentage change using the ‘pct_change’ function provided by the Pandas package into the ‘dc’ list. Then we created a dataframe named ftse100_momentum to store the symbol of the stock, percentage change, and the momentum which we calculated after this.

To calculate the momentum, we first created a column named ‘momentum’ in the ‘ftse100_momentum’ dataframe and filled it with null values. Then we passed on a for-loop to fill the null values with the actual momentum values.

Finding Stocks with Greater Momentum

First, we will create a dataframe named ‘top_picks’ which stores the top 10 stocks having greater momentum than the rest.
In order to sort and find the top 10 stocks, we used the ‘nlargest’ function provided by the Pandas package.

Backtesting the strategy

In this step, we are going to backtest Quantitative Momentum Strategy by investing money in the top 10 stocks with the highest momentum, and let’s see what results we get.

First, we will define two variables:

1. One to store the total investment value
2. And other to store the investment value per stock

Next, we will use a for loop to find the day’s close price of all the top 10 stocks and append those values into the ‘day_close’ variable. In further lines of code, we will create a dataframe named ‘backtest_df’ which will be comprised of the calculated number of stocks to buy in each of them, the returns we got by investing in those top 10 stocks, and finally the return percentage of our investment.

We have got some decent returns above!!

Conclusion

In this article, we learned about Quantitative Momentum Strategy and it's implementation in Python using DSS API to pick tradable stocks. We defined a function to authenticate with DSS Rest API and get a token, and further expanded FTSE ChainRic to get the list of all stocks of FTSE 100. We can consider stocks listed on other exchanges to experiment around as well. We defined another function to pull intraday data for FTSE ChainRic using DSS Rest API. By performing research given the amount of data being pulled, we can learn different ways of data extraction using DSS API along with extensive data processing. Further we can try tuning this strategy accordingly by using several other metrics.