In #DolphinDB we can import historical data into a stream table in chronological order as “real-time data” so that the same script can be used both for backtesting and real-time trading.
For more info, please check: https://medium.com/@DolphinDB_Inc/introduction-to-backtesting-strategy-historical-data-replay-in-dolphindb-497e24af596d.
#DolphinDB offers a unified batch and stream processing solution for machine learning, enabling efficient feature generation from both historical and real-time data. By constructing 10-minute features from 3-second snapshot data, the solution generates 676 derived features and proves to be 30 times faster than Python.
More info, please visit: https://medium.com/@DolphinDB_Inc/feature-engineering-for-stock-volatility-prediction-the-unified-stream-and-batch-processing-21bc38651b4.
This article introduces how to conduct feature engineering for model training and prediction in DolphinDB. Inspired by the 1st place solution of the time series prediction competition Kaggle Optiver…
Calculating transaction costs from tick data involves matching trade and NBBO tables, which have nanosecond-level timestamps that rarely align perfectly. #DolphinDB allows for complex, non-exact joins with just a single line of script, facilitating the calculation of transaction costs.
More info, please visit: https://medium.com/@DolphinDB_Inc/how-to-calculate-transaction-costs-in-dolphindb-228da46ee922.
#DolphinDB’s powerful functions and parallel computing capabilities enable quick and accurate correlation calculations, enhancing investment strategies and risk management.
More info, please visit: https://medium.com/@DolphinDB_Inc/mastering-pairwise-correlations-calculation-of-securities-through-coding-e840c8e8260a
#DolphinDB improves data analysis by optimizing query and aggregation calculations, offering fast and accurate data processing for large datasets. Through its distributed architecture and advanced techniques, #DolphinDB significantly accelerates data queries and aggregations, making it an essential tool for real-time data analysis across various industries.
More info, please visit: https://medium.com/@DolphinDB_Inc/accelerating-data-analysis-embracing-the-efficient-query-and-aggregation-calculations-c1dfb484d09b.
How to convert high-frequency signals into discrete buy/sell/hold signals？
It can be easily implemented in #DolphinDB with function iif！
More info, please visit: https://medium.com/@DolphinDB_Inc/high-frequency-data-analysis-converting-high-frequency-signals-to-discrete-buy-sell-signals-ea3146820424.
# DolphinDB exhibits outstanding performance in data downsampling due to the following reasons:
1. Jobs are executed distributedly and resources of different nodes can be utilized at the same time;
2. Compression reduces the disk I/O;
3. Columnar storage and vectorized computation improve the efficiency of aggregation.
More info, please visit: https://medium.com/@DolphinDB_Inc/how-to-downsample-your-data-efficiently-ef7125fb9a6.
Why is #DolphinDB’s code so concise and elegant? 😲
More info:
https://medium.com/@DolphinDB_Inc/a-simpler-way-to-calculate-worldquant-101-alphas-c55dac54e9f7.
Based on the #DolphinDB’s streaming replay, distributed database architecture and APIs, you can create a powerful tool for historical insight and model back-testing, which allows you to review situations of interest and improve future performance.
More info, please visit: https://medium.com/@DolphinDB_Inc/best-practices-for-market-data-replay-e69fea702736.
DolphinDB’s multi-source market data integration solution significantly enhances the cryptocurrency market data acquisition process, providing traders with a superior framework for strategic trading decisions and activities.
More info, please visit: https://medium.com/@DolphinDB_Inc/integrating-crypto-market-data-from-multiple-sources-a92d5ef5f264.
