数据驱动开发_开发数据驱动的股票市场投资方法

最新推荐文章于 2024-08-20 08:00:00 发布

翻译最新推荐文章于 2024-08-20 08:00:00 发布 · 285 阅读

0 ·

CC 4.0 BY-SA版权

原文链接：https://medium.com/swlh/developing-a-data-driven-approach-in-stock-market-investment-104c2b1e49c4

文章标签：

#python #java #人工智能 #大数据 #linux

本文介绍了一种数据驱动的投资方法，通过一次性分析所有股票而非单独分析每一只股票，避免了选择过程中的主观偏见。利用仪表板风格的可视化工具可以同时查看多个指标，帮助投资者快速做出更明智的决策。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

数据驱动开发

Data driven means that your decision are driven by data and not by emotions. This approach can be very useful in stock market investment. Here is a summary of a data driven approach which I have been taking recently

数据驱动意味着您的决定是由数据而非情感驱动的。这种方法在股票市场投资中非常有用。这是我最近采用的数据驱动方法的摘要

停止按库存分析库存 (Stop analysing stock by stock)

Being a data scientist, the one thing which I do not like about existing stock market tools is that the analysis is done in a very crude way. The analysis has still to be done for each stock symbol. Something like this

作为数据科学家，我不喜欢现有的股票市场工具的一件事就是分析是非常粗略的。必须对每个股票代号进行分析。像这样

Choose a stock, Analyse it, decide on buy/sell

选择股票，进行分析，决定购买/出售

Go to second stock, Analyse it, decide on buy/sell

转到第二只股票，进行分析，决定买入/卖出

Go to third stock, Analyse it, decide on buy/sell

转到第三只股票，进行分析，决定买入/卖出

etc…

等等…

Image for post — Analysing stock by stock — a very cumbersome way

This forces you to choose stock in first place. And the way you choose the stock is based on intuition or news you read about the stock. So there is already some bias introduced in the selection process.

这迫使您首先选择库存。选择股票的方式是基于您对股票的直觉或新闻。因此，在选择过程中已经引入了一些偏见。

The approach which I am using is top-down. Take all possible stocks and then analyse all together

我使用的方法是自上而下的。盘点所有可能的股票，然后一起分析

This way I do not have to think of selection of a particular stock in one place. I just take all stock symbols and analyse all together. Shown here is analysing all stock of CAC40 (French Stock Market)

这样，我不必在一个地方考虑选择特定股票。我只考虑所有股票代码并一起分析。此处显示的是分析CAC40(法国股票市场)的所有库存

What I get is graph of all stocks together.

我得到的是所有股票的图表。

This saves me immense amount of time. Earlier I used to spent hours to get the visualisation for 40 stocks. Now I can get visualisation for 40 stocks in matter of seconds

这为我节省了大量时间。早些时候，我曾经花了几个小时来获得40只股票的可视化。现在，我可以在几秒钟内获得40只股票的可视化

仪表板风格在同一视线中呈现多种视觉效果 (Dashboard style multiple visualisations in same sight)

For those who have been in data profession, know the importance of dashboards which gives multiple visualisations to see at the same time. It helps you make better decisions.

对于那些从事数据行业的人来说，了解仪表板的重要性是可以同时看到多个可视化内容的。它可以帮助您做出更好的决策。

This approach is very suitable for stock analysis, as one has to focus on multiple technical indicators.

这种方法非常适合于库存分析，因为它必须关注多个技术指标。

The above is a good example on power of “seeing” multiple visualisations at same time. The moving average shows an uptrend. However the candlestick graph shows loss in momentum. So there might be a downward trend which might be starting, so would be better to stay away from this stock

上面的示例很好地展示了同时“看到”多个可视化效果的能力。移动平均线显示上升趋势。但是，烛线图显示了动量损失。因此，可能会有开始下降的趋势，因此最好远离该库存

根据各种标准进行排序 (Sorting based on various criterions)

The sorting functionality is key for any data driven approach. So in my approach I have developed sorting based on various criterions such as Trend, Difference with Moving average etc..

排序功能是任何数据驱动方法的关键。因此，在我的方法中，我根据各种标准(例如趋势，与移动平均线的差异等)开发了排序。

Just by sorting based on Trend High to low, surfaces “safe or less risky” stocks at the top

仅根据趋势从高到低排序，顶部的“安全或风险较小”的库存

As you can see that CAP.PA surfaces to the top as it has been top trending since long time. This can be confirmed by its upward climbing moving average line. Now that is being data driven. Let data tell you the nature of the stock rather than deciding based on any intuitions

如您所见，CAP.PA一直处于顶部趋势，因为长期以来一直是趋势。这可以通过其向上攀升的移动平均线来确认。现在这是数据驱动的。让数据告诉您股票的性质，而不是根据任何直觉来决定