Home AI News Upgrade Your Algorithm: Harness the Power of Self-Serve Data

Upgrade Your Algorithm: Harness the Power of Self-Serve Data

Table of Contents:

Introduction
Benefits of Using Self-Serve Data
Example Algorithm Using Fetcher API
Uploading Data via Self-Serve Data
Integrating Self-Serve Data with Pipeline
Simplifying the Algorithm Code
Creating the Pipeline
Setting Constraints on Orders
Calculating an Optimal Portfolio
Recap and Conclusion

Updating Your Algorithm: A Guide to Using Self-Serve Data

1. Introduction

In this article, we will discuss the process of updating an algorithm that uses the fetcher API to instead utilize self-serve data in Quandl Peon. We will explore the benefits of self-serve data and provide a step-by-step example of adjusting an algorithm to make use of this feature.

2. Benefits of Using Self-Serve Data

Using self-serve data in Quandl Peon offers several advantages. Firstly, it seamlessly integrates your data into the platform's pipeline, ensuring there is no look-ahead bias. Additionally, self-serve data can be used alongside existing pricing fundamentals and alternative data available in pipeline. Lastly, strategies created with uploaded data are eligible for participation in the daily Quandl Peon contest and the allocation process.

3. Example Algorithm Using Fetcher API

Let's start by examining an algorithm that currently uses the fetcher API to read certain datasets. We will take inspiration from a community member's algorithm that applies the ds2 score and Asuran data set to make trading decisions. The algorithm defines an objective of maximizing alpha using the ds2 scores as weights, with constraints on position sizes and neutral allocation.

4. Uploading Data via Self-Serve Data

To make use of self-serve data, we need to upload the same data set without relying on the fetcher API. This can be done easily by utilizing the self-serve data feature in Quandl Peon. We'll walk you through the process of uploading the data set step-by-step, ensuring a smooth integration into your algorithm.

5. Integrating Self-Serve Data with Pipeline

Once the data set has been successfully uploaded via self-serve data, the next step is to load it into your algorithm using pipeline. We'll guide you through the process of importing pipeline and filtering the tradable stocks using the latest iteration of Quandl Peon's tradable Universe. This will ensure that only the stocks fitting your defined constraints are included in the pipeline output.

6. Simplifying the Algorithm Code

To fully utilize the uploaded data set, we need to update the algorithm code accordingly. We'll simplify the initialize function by removing the fetch CSV call and replacing it with the pipeline. Additionally, we'll modify the before trading start function to store the results of the daily pipeline in the strategies context variable, allowing easy access to the pipeline's results in other functions.

7. Creating the Pipeline

In order to incorporate the uploaded data set into our algorithm, we need to construct a pipeline using the ds2 scores as a pipeline column. We'll walk you through the process of creating the pipeline, including screening it with the tradable stocks filter and excluding any NaN ds2 score values. This will ensure that our algorithm makes use of the Relevant data in making trading decisions.

8. Setting Constraints on Orders

Before placing trades, it's important to constrain the orders to meet certain criteria. We'll guide you through setting three constraints: maximum gross exposure, dollar neutrality, and position concentration. These constraints aim to maintain the desired risk exposure and portfolio balance.

9. Calculating an Optimal Portfolio

To optimize the portfolio and achieve the defined objective subject to the constraints, we'll introduce the order optimal portfolio function. This function calculates the optimal portfolio using the alpha values from the pipeline. By utilizing this function, you can ensure your algorithm makes the most informed trading decisions based on the available data.

10. Recap and Conclusion

In this article, we've covered the process of updating an algorithm to use self-serve data in Quandl Peon. We started with the benefits of self-serve data, walked through an example algorithm using the fetcher API, and then demonstrated how to upload data via self-serve data. We then integrated the uploaded data with pipeline, Simplified the algorithm code, and set constraints on orders. Finally, we calculated an optimal portfolio using the order optimal portfolio function. We hope this guide has been helpful in understanding the steps involved in updating your algorithm.

Highlights:

Seamlessly integrate your own data into Quandl Peon's pipeline
Avoid look-ahead bias with self-serve data
Utilize uploaded data alongside existing pricing fundamentals and alternative data
Eligible for participation in the Quandl Peon contest and allocation process

FAQ: Q: What are the benefits of using self-serve data in Quandl Peon? A: Self-serve data integration, avoidance of look-ahead bias, and eligibility for contests and allocation process.

Q: Can I use self-serve data with existing pricing fundamentals and alternative data? A: Yes, self-serve data can be used alongside existing data in Quandl Peon's pipeline.

Q: How can I update my fetcher algorithm to use self-serve data? A: The article provides a step-by-step example of updating an algorithm to use self-serve data in Quandl Peon.

Resources: