An data analysis project
Budget: $240 – $2,000 HKD
This is a data analysis project. You may use whatever method you want to do. The goal is to use the data to predict the "cointegration_constant".
X (independent data): first 16 columns of the dataset
y : column "cointegration_constant"
Task: "milestone_date_o2" represents the end day of a quarter. We have to use the data before a specific milestone_date_o2 to predict the cointegration constant of this milestone_date
Example: For milestone_date_o2 20201231, we train the model (training data) with the data before 20201231 (20020630 to 20200930), that is row 0-19086 in the data set. The testing data is the row 19087-20591. The better the recall and precision, the better the result.
Goal: You need to use a consistent model structure (like same algorithm, same layers, same neuron etc) along all the milestone_date. After a model is built, train the model with the data described in "sample" section, from 20200930 to 20220331.
X (independent data): first 16 columns of the dataset
y : column "cointegration_constant"
Task: "milestone_date_o2" represents the end day of a quarter. We have to use the data before a specific milestone_date_o2 to predict the cointegration constant of this milestone_date
Example: For milestone_date_o2 20201231, we train the model (training data) with the data before 20201231 (20020630 to 20200930), that is row 0-19086 in the data set. The testing data is the row 19087-20591. The better the recall and precision, the better the result.
Goal: You need to use a consistent model structure (like same algorithm, same layers, same neuron etc) along all the milestone_date. After a model is built, train the model with the data described in "sample" section, from 20200930 to 20220331.