Using excel and it's macros function to gain an understanding of stock performances by year based on total trade volume and overall positive or negative performance reported by individual stocks. This project also is a teaching methodology to understand code (VBA) refactoring and how it can improve overall appearance and performance.
The stock funds dataset included two years of stock performance for only 12 publicly traded stocks. The two methods for determining best stock performance were total number of trades explained here as 'Total Volume', and overall growth in by percentage from the beginning of the year to the end of the year.
I wrote the code first for a specific stock "DQ" to understand macros performace and syntax. Then took it a step further and wrote the code to include all of the given stocks and their paired data resulting in a table that produced the total traded volume and the end of year gains or losses; formatted for easy viewing with green for gains and red for losses. The 2017 and 2018 performances show a stark comparasin when looking at the data parameters outlined; with 2017 out performing 2018 as a whole set. The only real standout stocke could be reportedly 'ENPH' for these two specific years. The other postitively performing one both years was 'RUN".
Some limitiations of this data set are only twelve stocks total, and the way they are being evaluated is wholly unlike the standard stock performance industry standardizations. However, for this learning module, the simplification makes more sense.
As you can see in Figures 1 & 2, the differences in code length, appearance, and in the following four figures (3:6); the refactored code was quicker. The main changes between the two were to take out a nested for loop. This iterates to a power of however many loops it has to go through which increases time by that factor. So the more iterations the more exponentially the time grows as the for loop essentially has to repeat actions. Using a better method takes care of that in the refactoring. Changes also include initializing the arrays so VBA does not have to automatically assign 'Long' as a default, which also takes up memory; creating a longer run time. Finally, the last change is to creat one variable that works throughout the refactor to index across the four different arrays, which enables us to use one for loop instead of nested loops.
The two time stamps while looking very close; still show how much faster the refactoring code works. Imagine a greater amount of data, the first code would have to at a multiplied factor take longer the more data it loops and loops over. It also is worth mentioning that running the first code on my computer does return a time that seems quick, but also locks my computer for about 20 seconds prior showing there is still processing going on with out me being able to capture that output.
- Figure 3. 2017 stock performance by total volume reported and gains or losses recorded, and time-stamped code-run time.
- *Figure 4. 2018 stock performance by total volume reported and gains or losses recorded, and time-stamped code-run time.
- Figure 5. 2017 stock performance by total volume reported and gains or losses recorded; REFACTORED time stamp.
- Figure 6. 2018 stock performance by total volume reported and gains or losses recorded; REFACTORED time stamp.
- What are the advantages and disadvantages of refactoring code in general?
These are the main advantages for refactoring in general, having a more efficient code, a code that is more useful in a broader sense, and making code easier to read. You are in essence making the code more accurate, while still usable. For example, when this code was first written there was no defining or labelling of some of the values, which left VBA to give its "best guess" or just assume all the values were Long data types taking up more memory than these values needed; thus slowing down the code performance. Added 'comments' makes the code easier to read, as well as explaining intent of that portion of the code. Should another coder take a look after you, they can easily understand where you were going with your code.
Some disadvantages of refactoring could be with more comments the code itself gets longer creating more opportunity for mis-steps. Another possible disadvantage is that the orginal coder may not know a way to refactor a part of their code in a more efficient manor. This is a key in working together with others, and seeking out different ways of syntax and answers, but still might remail ellusive for that particular code. A final potential downfall of refactoring is simply time. Depending on what the budget for the code looks like or the time parameters, this might be prohibitive, even after all its' advantages are so pronounced.
- What are the advantages and disadvantages of the original VBA Stock Analysis and the Refactored script?
These happen to be similar to some of the general advantages and disadvantages listed above. Namely, the code is easier to read, the comments help the code look neat, explain what each line is expecting and producing, and it definitely decreases overall run time, proving the refactoring is more efficient.
The disadvantages here are also mirroring the ones above, it took a longer amount of time to refactor some of this code based simply on my original lack of knowledge. Group sourcing solutions and 'plug-and-play' were key to figuring out the correct formulas to run this efficient code, but I still needed to expend the time to figure out which sources would work to the best advantage of this code. As I was not under budget constraints, this seems to not apply as much for this type of situation.