As a private investor, the sheer amount of information that can be found on the internet is rather daunting. Trying to understand what type of companies or ETFs are available is incredibly challenging with there being millions of companies amd derivatives available on the market. Sure, the most traded companies and ETFs can quickly be found simply because they are known to the public (for example, Microsoft, Tesla, S&P500 ETF or an All-World ETF). However, what else is out there is often unknown.
This database tries to solve that. It features 300.000+ symbols containing Equities, ETFs, Funds, Indices, Currencies, Cryptocurrencies and Money Markets. It therefore allows you to obtain a broad overview of sectors, industries, types of investments and much more.
The aim of this database is explicitly not to provide up-to-date fundamentals or stock data as those can be obtained with ease (with the help of this database) by using yfinance, FundamentalAnalysis or ThePassiveInvestor. Instead, it gives insights into the products that exist in each country, industry and sector and gives the most essential information about each product. With this information, you can analyse specific areas of the financial world and/or find a product that is hard to find. See for examples on how you can combine this database, and the earlier mentioned packages the section Examples.
Some key statistics of the database:
| Product | Quantity | Sectors | Industries | Countries | Exchanges |
|---|---|---|---|---|---|
| Equities | 155.705 | 16 | 242 | 111 | 82 |
| ETFs | 36.727 | 364* | 94* | 100** | 52 |
| Funds | 57.816 | 1678* | 438* | 100** | 34 |
| Product | Quantity | Category |
|---|---|---|
| Currencies | 2.590 | 174 Currencies |
| Cryptocurrencies | 3.624 | 299 Cryptocurrencies |
| Indices | 86.353 | 49 Exchanges |
| Money Markets | 1.384 | 2 Exchanges |
* These numbers refer to families (iShares, Vanguard) and categories (World Stock, Real Estate) respectively.
** This is an estimation. Obtaining the country distribution can only be done by collecting data on the underlying
or by manual search.
To access the database you can download the entire repository, but I strongly recommend making use of the package closely attached to the database. It allows you to select specific json files as well as search through collected data with a specific query.
You can install the package with the following steps:
pip install FinanceDatabase- Alternatively, download the 'Searcher' directory.
- (within Python)
import FinanceDatabase as fd
The package has the following functions:
show_options(product, equities_selection=None)- gives all available options from the functions below per product (i.e. Equities, Funds) which then can be used to collect data. You can select a sub selection of equities by entering 'countries', 'sectors' or 'industries' for equities_selection.select_cryptocurrencies(cryptocurrency=None)- with no input gives all cryptocurrencies, with input gives the cryptocurrency of choice.select_currencies(currency=None)- with no input gives all currencies, with input gives the currency of choice.select_etfs(category=None)- with no input gives all etfs, with input gives all etfs of a specific category.select_equities(country=None, sector=None, industry=None)- with no input gives all equities, with input gives all equities of a country, sector, industry or a combination of the three.select_funds(category=None)- with no input gives all funds, with input gives all funds of a specific category.select_indices(market=None)- with no input gives all indices, with input gives all indices of a specific market which usually refers to indices in a specific country (like de_market gives DAX).select_moneymarket(market=None)- with no input gives all moneymarkets, with input gives all moneymarkets of a specific market which usually refers to moneymarkets in a specific country.search_products(database, query, search='summary', case_sensitive=False, new_database=None)- with input from the above functions, this function searches for specific values (i.e. the query 'sustainable') in one of the keys of the dictionary (which is by default the summary). It also has the option to enable case-sensitive searching which is off by default.
For users of the broker DeGiro, you are able to find data on the tickers found in the
Commission Free ETFs list by selecting either
core_selection_degiro_filled (all data) or core_selection_degiro_filtered (filtered by summary) as category
when using the function select_etfs.
For additional information about each function you can use the build-in help function of Python. For
example help(show_options) returns a general description, the possible input parameters and what is returned
as output.
If you wish to store the database at a different location (for example your own Fork) you can do so with the variable
base_url which you can find in each of the above 'select' functions. An example would be:
select_funds(category='Africa Equity', base_url=<YOUR URL>)
You can also store the database locally and point to your local location with the variable base_url and by setting
use_local_location to True. An example would be:
select_etfs(category='Bank Loan', base_url='C:/Users/jerbo/FinanceDatabase/Database/ETFs/', use_local_location=True)
This section gives a few examples of the possibilities with this package. These are merely a few of the things you can do with the package. As you can obtain a wide range of symbols, pretty much any package that requires symbols should work.
Understanding which sectors exist in a country can be interesting. Not only to understand the focus of the country but
also to understand which area holds the most data. This is a demonstration of the show_options function.
A function crucial to querying data from the Database.
Let's start by acquiring the unique countries, sectors and industries of all equities in the database:
import FinanceDatabase as fd
# Obtain all countries from the database
equities_countries = fd.show_options('equities', 'countries')
# Obtain all sectors from the database
equities_sectors = fd.show_options('equities', 'sectors')
# Obtain all industries from the database
equities_industries = fd.show_options('equities', 'industries')
# Obtain all countries + sectors + industries from the database
equities_all_categories = fd.show_options('equities')
This gives the following lists (where equities_all_categories is a dictionary with these three lists):
Then, I want to see how many companies exist in each sector in the Netherlands. Let's count all companies with the following code, I skip a sector when it has no data and also do not include companies that are not categorized:
equities_per_sector_netherlands = {}
for sector in equities_sectors[1:]:
try:
equities_per_sector_netherlands[sector] = len(fd.select_equities(country='Netherlands', sector=sector))
except ValueError as error:
print(error)
Lastly, I plot the data in a pie chart and add some formatting to make the pie chart look a bit nicer:
legend, values = zip(*equities_per_sector_netherlands.items())
colors = ['b', 'g', 'r', 'c', 'm', 'y', 'k', 'tab:blue', 'tab:orange', 'tab:gray',
'lightcoral', 'yellow', 'saddlebrown', 'lightblue', 'olive']
plt.pie(values, labels=legend, colors=colors,
wedgeprops={'linewidth': 0.5, 'edgecolor': 'white'})
plt.title('Companies per sector in the Netherlands')
plt.tight_layout()
plt.show()
This results in the following graph which gives an indication which sectors are dominant within The Netherlands. Of course this is a mere example and to truly understand the importance of certain companies for the Netherlands, an in-depth analysis must be done.
With the help of ta and yfinance I can quickly perform a basic technical analysis on a group of ETFs categorized by the FinanceDatabase. I start by searching the database for ETFs related to Health and then make a subselection by searching, in the collected database, for biotech-related ETFs:
import FinanceDatabase as fd
health_etfs = fd.select_etfs(category='Health')
health_etfs_in_biotech = fd.search_products(health_etfs, 'biotech')
Then, I collect stock data on each ticker and remove tickers that have no data in my chosen period. The period I have chosen shows the initial impact of the Coronacrisis on the financial markets.
import yfinance as yf
stock_data_biotech = yf.download(list(health_etfs_in_biotech.keys()), start="2020-01-01", end="2020-06-01")['Adj Close']
stock_data_biotech = stock_data_biotech.dropna(axis='columns')
Next up I initialise subplots and loop over all collected tickers. Here, I create a new temporary DataFrame that I fill with the adjusted close prices of the ticker as well as the Bollinger Bands. Then I plot the data in one of the subplots.
import pandas as pd
from ta.volatility import BollingerBands
import matplotlib.pyplot as plt
figure, axis = plt.subplots(4, 3)
row = 0
column = 0
for ticker in stock_data_biotech.columns:
data_plot = pd.DataFrame(stock_data_biotech[ticker])
indicator_bb = BollingerBands(close=stock_data_biotech[ticker], window=20, window_dev=2)
data_plot['bb_bbm'] = indicator_bb.bollinger_mavg()
data_plot['bb_bbh'] = indicator_bb.bollinger_hband()
data_plot['bb_bbl'] = indicator_bb.bollinger_lband()
axis[row, column].plot(data_plot)
axis[row, column].set_title(health_etfs_in_biotech[ticker]['long_name'], fontsize=6)
axis[row, column].set_xticks([])
axis[row, column].set_yticks([])
column += 1
if column == 3:
row += 1
column = 0
figure.suptitle('Technical Analysis of Biotech ETFs during Coronacrisis')
figure.tight_layout()
This leads to the following graph which gives an indication wether Biotech ETFs were oversold or overbought and how this effect is neutralised (to some degree) in the months after. Read more about Bollinger Bands here.
If I wish to obtain all companies within the United States listed under 'Airlines' I can write the following code:
import FinanceDatabase as fd
airlines_us = fd.select_equities(country='United States', industry='Airlines')
Then, I can use packages like yfinance to quickly collect data from Yahoo Finance for each symbol in the industry like this:
from yfinance.utils import get_json
from yfinance import download
airlines_us_fundamentals = {}
for symbol in airlines_us:
airlines_us_fundamentals[symbol] = get_json("https://finance.yahoo.com/quote/" + symbol)
airlines_us_stock_data = download(list(airlines_us))
With a few lines of code, I have collected all data from a specific industry within the United States. From here on you can compare pretty much any key statistic, fundamental- and stock data. For example, let's plot a simple bar chart that gives insights in the Quick Ratios (indicator of the overall financial strength or weakness of a company):
import matplotlib.pyplot as plt
for symbol in airlines_us_fundamentals:
quick_ratio = airlines_us_fundamentals[symbol]['financialData']['quickRatio']
long_name = airlines_us_fundamentals[symbol]['quoteType']['longName']
if quick_ratio is None:
continue
plt.barh(long_name, quick_ratio)
plt.tight_layout()
plt.show()
Which results in the graph displayed below (as of the 14th of June 2021). From this graph you can identify companies that currently lack enough assets to cover their liabilities (quick ratio < 1), and those that do have enough assets (quick ratio > 1). Both too low and too high could make you wonder whether the company adequately manages its assets.
If I want to understand which listed technology companies exist in Silicon Valley, I can collect all equities of the sector 'Technology' and then filter based on city to obtain all listed technology companies in 'Silicon Valley'. The city 'San Jose' is where Silicon Valley is located. I remove all tickers with a dot since they refer to different markets.
import FinanceDatabase as fd
all_technology_companies = fd.select_equities(sector='Technology')
silicon_valley = fd.search_products(all_technology_companies, query='San Jose', search='city')
for ticker in silicon_valley.copy():
if '.' in ticker:
del silicon_valley[ticker]
Then I start collecting data with the FundamentalAnalysis package. Here I collect the key metrics which include 57 different metrics (ranging from PE ratios to Market Cap).
import FundamentalAnalysis as fa
API_KEY = "YOUR API KEY HERE"
data_set = {}
for ticker in silicon_valley:
try:
data_set[ticker] = fa.key_metrics(ticker, API_KEY, period='annual')
except Exception:
continue
Then I make a selection based on the last 5 years and filter by market cap to compare the companies in terms of size with each other. This also causes companies that have not been listed for 5 years to be filtered out of my dataset. Lastly, I plot the data.
import pandas as pd
import matplotlib.pyplot as plt
years = ['2016', '2017', '2018', '2019', '2020']
market_cap = pd.DataFrame(index=years)
for ticker in data_set:
try:
data_years = []
for year in years:
data_years.append(data_set[ticker].loc['marketCap'][year])
market_cap[all_technology_companies[ticker]['short_name']] = data_years
except Exception:
continue
market_cap_plot = market_cap.plot.bar(stacked=True, rot=0, colormap='Spectral')
market_cap_plot.legend(prop={'size': 5.25})
plt.show()
This results in the graph displayed below which separates the small companies from the large companies. Note that this does not include all technology companies in Silicon Valley because most are not listed or are not included in the database of the FundamentalAnalysis package.
Sometimes, Excel simply offers the best solution if you want compare a range of ETFs quickly. Therefore, another option is to use my program ThePassiveInvestor. The goal of this program is to quickly compare a large selection of ETFs by collecting their most important attributes (i.e. holdings, return, volatility, tracking error).
As I invest with DeGiro, a great start for me would be by collecting all ETFs that are listed within the Core Selection (commission free) list of my broker with the following code (or manually obtain them from the json file):
import FinanceDatabase as fd
core_selection = fd.select_etfs("core_selection_filtered")
Then I convert the keys of the core_selection into a Series and send it to Excel without index and header.
import pandas as pd
tickers = pd.Series(core_selection.keys())
tickers.to_excel('core_selection_tickers.xlsx', index=None, header=None)
If you open the Excel file created you see the following lay-out (which corresponds to the lay-out accepted by the program):
Then I open ThePassiveInvestor program and use the Excel as input. The first input is the Excel that you want to be filled with input from your tickers (created by the program). The second input is the file you created above.
When you run the program it starts collecting data on each ticker and fills the Excel with data. After the program is finished you are able to find an Excel that looks very much like the GIF you see below. With this data you can get an indication whether the ETF is what you are looking for.
In this section you can find answers to commonly asked questions. In case the answer to your question is not here, consider creating an Issue.
- How did you get your data?
- Please check the Methodology.
- Is there support for <insert_country>?
- Yes, most likely there is as the database includes 111 countries. Please check here.
- How can I find out which countries, sectors and/or industries exists within the database without needing to check
the database manually?
- For this you can use the
show_optionsfunction from the package attached to this database. Please see this example
- For this you can use the
- When I collect data via yfinance I notice that not all tickers return output, why is that?
- Some tickers are merely holdings of companies and therefore do not really have any data attached to them. Therefore, it makes sense that not all tickers return data. If you are still in doubt, search the ticker on Google to see if there is really no data available.
- How frequently does the Database get updated?
- I aim at doing this every few months. The database does not have to get updated frequently because the data collected is only general information. For example, a Sector name hardly changes and companies do not tend to move to another country every few months. Therefore, the data should stay up to date for several months. If you wish to contribute to updating the database then that is much appreciated. Please check the Methodology for guidance on how.
- Do your sector and industry names use the same naming convention as GIC sector?
- Not entirely but very similar, it's based on Yahoo Finance's sectors and industries. See industries and sectors. Perhaps a future adjustment could be to make them aligned with GICS.
Projects are bound to have (small) errors and can always be improved. Therefore, I highly encourage you to submit issues and create pull requests to improve the package.
The last update to the database is the 14th of June 2021. I always accept Pull Requests every few months to keep the database up to date. Extending the amount of tickers and data is also much appreciated. When you wish to do this, please inform me first to prevent multiple users doing the exact same thing.
