Icon

01_​Stocks

Stock Analysis: Automatic value prediction and alertingThe input to this workflow is a list of stock symbols from Yahoo Finance. Afer some preprocessing to determine the date range, this information is sent to the pandas webreader library (through the Python Source node) to collect(near) realtime stock information.The information is sent to the modeling workflow (through the Call Workflow node) where feature creation, model estimation, and prediction takes place for each stock symbol. This workflow collects the predictions and thereliability of the model, and visually displays the results - the expected percentage change in value for each stock symbol.Note: To use this workflow you need to have Python and the following libraries installed on your machine:-pandas as pd-pandas-datareader-datetime This section creates the workflow variables that are sent to the Python Source node The workflow variables are sent to the Python Source node, which gathers historical data for each symbol. It can occur that data for a specific symbol is (temporarily) unavailable, in which case a value of -99 is returned.These cases are sent to the bottom port of the If Switch node and are filtered out after the Loop End node. The predicted increased in stock value is sorted and exported to a plot and .csv files.When deployed on KNIME server, the Send Email node can be used to alert traders of potentialinteresting stock developments on a daily basis. Authentication is supported via workflow or node levelcredentials; connection security via STARTTLS and SSL are both supported. Extract today's dateDeduct one dayCollect stockdata from Yahoo FinanceConvert to workflow variablesDecompose date to day, month and yearExtract date field from RowIDInitiate a loop for each stock symbolMerge date variableswith stock symbolAttach .csv file andplot of stock predictions to an e-mailSort symbols by expected% increaseGather all predictionsFilter unavailablesymbolsWrite daily predictions to a .csvCall modelling workflowExport plot to a fileUnavailable Symbolsare set to -99Add predictionsto history fileInsert a list of stock symbols Create Date&TimeRange Date&Time Shift Python Source Table Rowto Variable Extract Date&TimeFields RowID Table Row ToVariable Loop Start Merge Variables Send Email Sorter IF Switch End IF Loop End Rule-basedRow Filter CSV Writer Call Workflow(Table Based) Image Writer (Port) Set format forunavailable stocks CSV Writer Determine stockavailability Table Creator Bar Chart Stock Analysis: Automatic value prediction and alertingThe input to this workflow is a list of stock symbols from Yahoo Finance. Afer some preprocessing to determine the date range, this information is sent to the pandas webreader library (through the Python Source node) to collect(near) realtime stock information.The information is sent to the modeling workflow (through the Call Workflow node) where feature creation, model estimation, and prediction takes place for each stock symbol. This workflow collects the predictions and thereliability of the model, and visually displays the results - the expected percentage change in value for each stock symbol.Note: To use this workflow you need to have Python and the following libraries installed on your machine:-pandas as pd-pandas-datareader-datetime This section creates the workflow variables that are sent to the Python Source node The workflow variables are sent to the Python Source node, which gathers historical data for each symbol. It can occur that data for a specific symbol is (temporarily) unavailable, in which case a value of -99 is returned.These cases are sent to the bottom port of the If Switch node and are filtered out after the Loop End node. The predicted increased in stock value is sorted and exported to a plot and .csv files.When deployed on KNIME server, the Send Email node can be used to alert traders of potentialinteresting stock developments on a daily basis. Authentication is supported via workflow or node levelcredentials; connection security via STARTTLS and SSL are both supported. Extract today's dateDeduct one dayCollect stockdata from Yahoo FinanceConvert to workflow variablesDecompose date to day, month and yearExtract date field from RowIDInitiate a loop for each stock symbolMerge date variableswith stock symbolAttach .csv file andplot of stock predictions to an e-mailSort symbols by expected% increaseGather all predictionsFilter unavailablesymbolsWrite daily predictions to a .csvCall modelling workflowExport plot to a fileUnavailable Symbolsare set to -99Add predictionsto history fileInsert a list of stock symbols Create Date&TimeRange Date&Time Shift Python Source Table Rowto Variable Extract Date&TimeFields RowID Table Row ToVariable Loop Start Merge Variables Send Email Sorter IF Switch End IF Loop End Rule-basedRow Filter CSV Writer Call Workflow(Table Based) Image Writer (Port) Set format forunavailable stocks CSV Writer Determine stockavailability Table Creator Bar Chart

Nodes

Extensions

Links