{"id":46624,"date":"2020-12-16T00:00:00","date_gmt":"2020-12-16T08:00:00","guid":{"rendered":"https:\/\/griddb-linux-hte8hndjf8cka8ht.westus-01.azurewebsites.net\/blog\/stock-market-analysis-with-python-pandas-plotly-and-griddb\/"},"modified":"2025-11-13T12:55:06","modified_gmt":"2025-11-13T20:55:06","slug":"stock-market-analysis-with-python-pandas-plotly-and-griddb","status":"publish","type":"post","link":"https:\/\/www.griddb.net\/en\/blog\/stock-market-analysis-with-python-pandas-plotly-and-griddb\/","title":{"rendered":"Stock Market Analysis with Python Pandas, Plotly and GridDB"},"content":{"rendered":"

Introduction <\/h2>\n

Stock markets are fickle and often changing. Humans have tried to tame the bull throughout history but have never been successful. Stock market prediction is difficult because there are too many factors at play, and creating models to consider such variances is almost impossible. However, recent advances in machine learning and computing have allowed machines to process large amounts of data. This will enable us to use past stock exchange data and analyze trends. This post will leverage python and GridDB to analyze stock data for Google for the past year.<\/p>\n

Stock prices are stored daily. Thus, daily stock data can grow very large. We will use GridDB as the database to store our data as it has been known to handle large datasets well. GridDB ensures high performance while being scalable and reliable at the same time. GridDB is easy to use and works with most modern programming languages. In this post, we will run some fundamental stock price analysis using python and visualize our results.<\/p>\n

FULL SOURCE CODE <\/a><\/p>\n

Setup<\/h2>\n

GridDB setup<\/h3>\n

Setup for the GridDB python client can be found in this video<\/a>. Before we begin, we should have the following endpoints correctly defined.<\/p>\n

\n
export LIBRARY_PATH=$LIBRARY_PATH<\/span>:\/usr\/share\/doc\/griddb-c-client<\/span> [insert path to c_client]\nexport PYTHONPATH=$PYTHONPATH<\/span>:<\/span>[insert path to python_client]\nexport LIBRARY_PATH=$LD_LIBRARY_PATH<\/span>:<\/span>[insert path to c_client\/bin]\n<\/code><\/pre>\n<\/div>\n
Python Libraries<\/h4>\n
We will use python 3.6<\/code> for this analysis. To install libraries we will use pip<\/code>.<\/p>\n
\n
pip install<\/span> pandas\npip install<\/span> plotly\n<\/code><\/pre>\n<\/div>\n
Data Collection<\/h3>\n
GridDB provides an excellent interface to access data. The GridDB python client blog goes into great detail on linking a GridDB database and pushing all the data to a pandas data frame. We will use yahoo finance to get data for Google stock. The data can be found at : Yahoo! Finance<\/a> We save the data for one year at GOOG.csv<\/code>.<\/p>\n
We can insert and retrive this data into GridDB with SQL queries. <\/p>\n
To insert:<\/p>\n
\n
import<\/span> griddb_python as griddb\nimport<\/span> pandas as pd\n\n# Initialize container<\/span>\ngridstore<\/span> = factory.get_store(host=<\/span> host, port=port,<\/span> \n            cluster_name=cluster_name,<\/span> username=uname,<\/span> \n            password=pwd)<\/span>\n\nconInfo<\/span> = griddb.ContainerInfo(\"GOOGL\"<\/span>,\n                    [[\"Date\"<\/span>, griddb.Type.TIMESTAMP],\n                    [\"Open\"<\/span>,griddb.Type.LONG],\n                    [\"High\"<\/span>, griddb.Type.LONG],\n                    [\"Low\"<\/span>,griddb.Type.LONG]\n                    [\"Close\"<\/span>, griddb.Type.LONG]\n                    [\"Adj. Close\"<\/span>, griddb.Type.LONG]\n                    [\"Volume\"<\/span>, griddb.Type.LONG]],\n                    griddb.ContainerType.COLLECTION, True)\n\ncont<\/span> = gridstore.put_container(conInfo)    \ncont.create_index(\"Date\"<\/span>, griddb.IndexType.DEFAULT)\ndata<\/span> = pd.read.csv(\"GOOG.csv\"<\/span>)\n#Add data<\/span>\nfor i in<\/span> range(len(data)):\n    ret<\/span> = cont.put(data.iloc[i, :])\n<\/code><\/pre>\n
Note that this process can be automa,ted to query regualrly from Yahoo! Finacne as well.
\nWe can retrive data from GridDB using the following SQL query:<\/p>\n
\n
query = cont.query(\"select<\/span> * where<\/span> Date<\/span> > TIMESTAMPADD<\/span>(YEAR<\/span>, NOW<\/span>(), -1<\/span>)\")<\/span>\n<\/code><\/pre>\n<\/div>\n
Data Analysis<\/h3>\n
We can load the data using pandas<\/code><\/p>\n
\n
Import<\/span> pandas as<\/span> pd\ngoogf = pd.read_csv(\"GOOG.csv\"<\/span>)\n<\/code><\/pre>\n<\/div>\n
This is what stock data looks like:
\n\"\"
\nThe columns are interpreted as:<\/p>\n