Deep Learning Application Involving Canadian Economic Data
Budget: €20 – €50 EUR
Deep Learning Application Involving Canadian Economic Data
-- Data set: about 30 years of monthly economic data for Canada, with about 30 variables for each month (Gross National Product, housing starts, workforce size, the interest rate, inflation, etc.)
-- We want the machine to determine the relationships between the variables and predict the next 12 months for each variable.
-- It would appear from the results published by others that LSTM algorithms are best for this project.
-- Now comes the fun part: for the coming 12 months, we want to play with the values of some of the independent variables and then see the results. For example, if we set the interest rate at (say) 4%, how would it affect all the other variables in the coming 12 month period.
-- In other words, we want to force the values of some of the data to the future periods and then let the algorithm calculate all the other values. Is this even possible to do with LSTM or any other deep learning algorithm?
-- If not, how best to find a programmer who I might hire to adjust the LSTM algorithm to allow certain data to be inserted in hidden layers?
-- Data set: about 30 years of monthly economic data for Canada, with about 30 variables for each month (Gross National Product, housing starts, workforce size, the interest rate, inflation, etc.)
-- We want the machine to determine the relationships between the variables and predict the next 12 months for each variable.
-- It would appear from the results published by others that LSTM algorithms are best for this project.
-- Now comes the fun part: for the coming 12 months, we want to play with the values of some of the independent variables and then see the results. For example, if we set the interest rate at (say) 4%, how would it affect all the other variables in the coming 12 month period.
-- In other words, we want to force the values of some of the data to the future periods and then let the algorithm calculate all the other values. Is this even possible to do with LSTM or any other deep learning algorithm?
-- If not, how best to find a programmer who I might hire to adjust the LSTM algorithm to allow certain data to be inserted in hidden layers?