This repository was made as part of my application to the company Next Gate Tech. It includes my take on the technical test that was sent to me on February 28th, 2020. The test's description can be found under others/NGT - Test.pdf The actual files answering the questions can be found in the folder case_study.
- Task A: Back-end can be found in the folder case_study/backend.
- Task B: Front-end can be found in the folder case_study/frontend.
- Task C: Cloud Functions can be found in the folder case_study/cloud.
Some other folders can also be found in this repository:
- case_study/data that contains the data used in the project, including the given csv file given with the test (example_data.csv).
- case_study/identification that contains the files for identification with the Google Cloud Plateform.
- others contains different files that are not relevant to the project but that I included for more transparency on my take on this test.
All rights of the technical test belong to Next Gate Tech S.A..
This part was answered using a jupyter notebook that allowed for an easier data exploring, comprehension and communication with Firestore.
The jupyter notebook already contains step by step explanation on the process.
Python 3.7 and the library pandas were used. The dataset had the following schema, I also included my interpretation of the columns.
| Column name | Interpretation |
|---|---|
| Subfund_Code | Subfund_Code |
| Share_Class_Code | Share_Class_Code |
| Valuation_Date | Date of the valuation of the different variables of the following columns |
| Subfund_CCY | Currency Code used for the Subfund |
| Net_Asset_Value | Value of the asset as of Valuation_Date's date |
| Nb_Shares_Outstanding | All shares currently owned by stockholders, company officials, and investors in the public domain |
| NAV_Per_Share | Net Asset Value Per Share |
| CCY_NAV_share | Currency code used for the NAV_Per_Share |
| ISIN_Code | International Securities Identification Number |
| Nb_Shares_Subscribed | Number of shares subscribed for this instrument on Valuation_Date's date |
| Amount_Subscribed | Nb_Shares_Subscribed x Net_Asset_Value |
| Nb_Shares_Redeemed | Number of shares redeemed for this instrument on Valuation_Date's date |
| Amount_Redeemed | Nb_Shares_Redeemedx Net_Asset_Value |
| Subfund_Long_Name | Always NaN |
| Dividend_Per_Share | Dividend that the fund should receive per share |
| Dividend_Payment_Date | Date at which the dividends should be paid |
| Dividend_PCCY | Always 0 |
| Dividend_Payment_CCY | Currency code of the dividend payment |
| Dividend_Subfund_CCY | Currency code of the dividend subfund |
| Total_Net_Assets | Total net assets of this instrument |
The following checks were made:
- There is only one row per instrument and valuation date.
- For each valuation date NAV_Per_Share=Total_Net_Assets/Nb_Shares_Outstanding
- For each valuation date Amount_Subscribed / Nb_Shares_Subscribed=Amount_Redeemed / Nb_Shares_Redeemed
We also calculated the correlation for each instruments between each other, here is the pearson correlation matrix of the different instruments:
To gain a bit of time for this part, I have used an already existing open source Angular 8 template that I have simplified and adapted to my own usage. The template used is black-dashboard-angular (under MIT License).
The front-end after integration of the firebase data looks like this:
This last task is so that the code is online but it can also be found in the folder case_study/cloud with a requirements.txt and a aggregate_subscribed_redeemed_checks.py file which is the name of the function on the Google Cloud Plateform.
Here's the function:
The trigger was set as asked but sadly it does not seem to work when the collection is being created even though it seems as if the trigger was set correctly:
Maybe the problem is that the functionality is still in bêta or maybe I did not choose the right trigger.* On the other hand, the cloud function does work on testing.