Due Date:November 3rd, 2024
Submission Requirements:
● Python code file(.py) containing all functions and logic.
● Excel workbookwith crypto historical data.
● Short project explanationin a text document.
● 2-minute Loom videoexplaining your approach.
● Submission Email:[email protected]
Internship Details:
● Stipend: €200-400/month
● Duration: 3 months
● Location: Remote
Assessment Task: Crypto Historical Data Retrieval
his task requires you to retrieve historical trading data for frequently traded cryptocurrency pairs and
T
calculate several analytical metrics. The main focus is on working with APIs, handling and processing data,
and calculating trading metrics over specific timeframes.
Task Breakdown and Detailed Requirements
Step 1: Identify a Reliable API for Crypto Data Retrieval
1. API Research and Selection
○ Find areliable, free APIthat provideshistorical datafor cryptocurrencies.
○ Ensure the API can fetch daily historical data fortop-traded cryptocurrency pairs.
○ Verify and documentthe following API details:
■ Number of crypto pairs supported: List available pairs if possible.
■ A
vailable timeframes: Confirm that daily, hourly, and/or weekly timeframes are
supported.
■ D
ata availability range: Earliest available date and most recent date for data
retrieval.
Step 2: Retrieve Historical Data
1. Create a Python Function for Data Retrieval
○ Develop a function namedfetch_crypto_datathat:
■ A
ccepts two inputs:crypto_pair(e.g., "BTC/USD") andstart_date(inYYYY-MM-DD
format).
■ R
etrievesdaily historical datafor the specified cryptocurrency pair starting from
start_date.
Breakout Consultancy Private Limited
CO: Neue Rothofstraße 13-19, 60313 - Frankfurt am Main, Germany | CIN: U70200UT2024PTC017145 | E:
[email protected] |
www.breakoutinvesting.in | Tel: +49 151 6858 3271
■ Returns aDataFramewith the following columns:
■ Date: The date of each record.
■ Open: Opening price of the cryptocurrency.
■ High: Highest price of the cryptocurrency.
■ Low: Lowest price of the cryptocurrency.
■ Close: Closing price of the cryptocurrency.
Step 3: Calculate Highest, Lowest, and Percentage Difference Metrics
1. Create a Python Function for Calculating Metrics
○ Name this functioncalculate_metrics. This function should:
■ Accept the following inputs:
■ data: The DataFrame containing historical crypto data.
■ v
ariable1: Integer representing a look-back period (e.g., 7 days) for
historical high and low metrics.
■ v
ariable2: Integer representing a look-forward period (e.g., 5 days) for
future high and low metrics.
2. Calculate and Add Columns as Below:
○ Historical High Price(Column:High_Last_{variable1}_Days)
■ Calculate the highest price in the past{variable1}days.
○ Days Since High(Column:Days_Since_High_Last_{variable1}_Days)
■ C
alculate the difference in days between the currentDateand the date of the
High_Last_{variable1}_Days.
○ % Difference from Historical High(Column:%_Diff_From_High_Last_{variable1}_Days)
■ C
alculate the percentage difference between the currentCloseprice and the high
price found in the past{variable1}days.
○ Historical Low Price(Column:Low_Last_{variable1}_Days)
■ Calculate the lowest price in the past{variable1}days.
○ Days Since Low(Column:Days_Since_Low_Last_{variable1}_Days)
■ C
alculate the difference in days between each date and the date of the calculated
low.
○ % Difference from Historical Low(Column:%_Diff_From_Low_Last_{variable1}_Days)
■ C
alculate the percentage difference between the currentCloseprice and the low
price found in the past{variable1}days.
○ Future High Price(Column:High_Next_{variable2}_Days)
■ Calculate the highest price within the next{variable2}days.
○ % Difference from Future High(Column:%_Diff_From_High_Next_{variable2}_Days)
Breakout Consultancy Private Limited
CO: Neue Rothofstraße 13-19, 60313 - Frankfurt am Main, Germany | CIN: U70200UT2024PTC017145 | E:
[email protected] |
www.breakoutinvesting.in | Tel: +49 151 6858 3271
■ C
alculate the percentage difference between the currentCloseprice and the value
fromHigh_Next_{variable2}_Days.
○ Future Low Price(Column:Low_Next_{variable2}_Days)
■ Calculate the lowest price within the next{variable2}days.
○ % Difference from Future Low(Column:%_Diff_From_Low_Next_{variable2}_Days)
■ C
alculate the percentage difference between the currentCloseprice and the value
fromLow_Next_{variable2}_Days.
Sample Data (Transposed to fit on the page)
024-01 2
2 024-01 2
024-01 2
024-01 2
024-01 2
024-01 2
024-01 2
024-01
Date -01 -02 -03 -04 -05 -06 -07 -08
Open 50000 50500 51500 52500 53000 53500 54000 54500
High 51000 52000 53000 53500 54000 54500 55000 55500
Low 49500 50000 50500 51000 51500 52000 52500 53000
Close 50500 51500 52500 53000 53500 54000 54500 55000
High_Last_7_Days 51000 52000 53000 53500 54000 54500 55000 55500
ays_Since_High_La
D
st_7_Days 1 1 1 1 1 1 1 1
_Diff_From_High_L
%
ast_7_Days -1.96% -0.96% -0.94% -0.93% -0.93% -0.92% -0.91% -0.90%
Low_Last_7_Days 49500 49500 50000 50500 51000 51500 52000 52500
ays_Since_Low_Las
D
t_7_Days 2 2 2 3 4 5 6 7
_Diff_From_Low_La
%
st_7_Days 2.02% 4.04% 5.00% 4.95% 4.90% 4.85% 4.81% 4.76%
High_Next_5_Days 52000 53000 53500 54000 54500 55000 55500 56000
_Diff_From_High_N
%
ext_5_Days 2.97% 2.91% 1.90% 1.89% 1.87% 1.85% 1.80% 1.78%
Breakout Consultancy Private Limited
CO: Neue Rothofstraße 13-19, 60313 - Frankfurt am Main, Germany | CIN: U70200UT2024PTC017145 | E:
[email protected] |
www.breakoutinvesting.in | Tel: +49 151 6858 3271
Low_Next_5_Days 49000 50000 51000 51500 52000 52500 53000 53500
_Diff_From_Low_Ne
%
xt_5_Days -2.97% -2.91% -2.86% -2.83% -2.81% -2.78% -2.75% -2.73%
3. Machine Learning Model
In this section, you will create a machine learning model that predicts future price differences based
on recent historical data. This additional component will assess your ability to build and evaluate a
machine learning model using calculated trading metrics.
1. Objective:
○ Train a machine learning model to predict two target variables:
■ %_Diff_From_High_Next_{variable2}_Days
■ %_Diff_From_Low_Next_{variable2}_Days
○ Use the following four features as input variables:
■ Days_Since_High_Last_{variable1}_Days
■ %Diff_From_High_Last{variable1}_Days
■ Days_Since_Low_Last_{variable1}_Days
■ %Diff_From_Low_Last{variable1}_Days
2. Requirements:
○ C
reate a separate Python file,ml_model.py, that contains the code for training and
testing this machine learning model.
○ Implement two main functions inml_model.py:
■ train_model: This function will train the machine learning model, evaluate its
performance, and return the model’s accuracy on a test dataset.
■ p
redict_outcomes: This function will use the trained model to make
predictions based on new input values for the four features. The function
should take the values of these four features as input and output predicted
values for%_Diff_From_High_Next_{variable2}_Daysand
%_Diff_From_Low_Next_{variable2}_Days.
Additional Requirements and Documentation
1. Documentation
○ Code Comments: Clearly document each function and major calculation step.
○ S
hort Project Explanation: Provide a text document summarizing your approach,
decisions, and challenges faced.
2. Loom Video: Record a 2-minute Loom video explaining:
Breakout Consultancy Private Limited
CO: Neue Rothofstraße 13-19, 60313 - Frankfurt am Main, Germany | CIN: U70200UT2024PTC017145 | E:
[email protected] |
www.breakoutinvesting.in | Tel: +49 151 6858 3271
○ Your approach to solving the task.
○ Any challenges faced and how they were overcome.
○ How you utilized any AI tools (if applicable).
Deliverables
● Python Code File:
○ Code for API data retrieval and metric calculations should be modular and well-documented.
● Excel Workbook:
○ Include sheets for each cryptocurrency pair’s historical data.
● Short Project Explanation:
○ Provide insights into your approach, challenges, and rationale.
● Loom Video:
○ Explain your process, approach, and any unique solutions you applied.
Evaluation Criteria
1. API Selection and Data Retrieval: Ability to research, choose, and use a suitable API.
2. Data Handling and Excel Output: Skill in structuring, formatting, and exporting data effectively.
3. Functionality and Accuracy: Ability to calculate metrics correctly and efficiently.
4. Code Clarity and Documentation: Clear organization and readable code.
5. Communication in Loom Video: Ability to clearly and confidently explain your approach.
Next Steps
If we are impressed with your submission, we will invite you for an interview to discuss your approach,
thought process, and coding decisions in more detail.
Breakout Consultancy Private Limited
CO: Neue Rothofstraße 13-19, 60313 - Frankfurt am Main, Germany | CIN: U70200UT2024PTC017145 | E:
[email protected] |
www.breakoutinvesting.in | Tel: +49 151 6858 3271
