How can I use Genetic Algoritm to predict future cost? I am stuck and would like some guidance to the next step

% Import & initialize gold project data
% from file:
% Training_gold.xlsx
T = readtable ("Testing gold (23-7-2019).xlsx");
% Specify data range used as input
DataRange = "B5:J14";

% Specify column names and types
VariableNames = ["gold", "shcap", "shunit", "truckcap", "truckunit", "eqltime", "midevtime", "miprod", "capex"];
SelectedVariableNames = ["gold", "shcap", "shunit", "truckcap", "truckunit", "eqltime", "midevtime", "miprod", "capex"];
VariableTypes = ["double", "double", "double", "double", "double", "double", "double", "double", "double"];

% Create the variables
goldprice = ('gold');
shovelcap = ('shcap');
shovelunit = ('shunit');
truckcap = ('truckcap');
truckunit = ('truckunit');
eqltime = ('eqltime');
midevtime = ('minedevtime');
miprod = ('miprod');
capex = ('capex');

%Save the variables into one *.mat file

save Training_gold_1 goldprice shovelcap shovelunit truckcap truckunit eqltime midevtime miprod capex 

%Clear them out of the workspace
clear goldprice shovelcap shovelunit truckcap truckunit eqltime midevtime miprod capex

%Load the cost variables
load ("Training_gold_1.mat");

Answer 1

[Star Strider]

The ga (Genetic Algorithm) function is an optimisation function. It will provide the best parameter estimates for the mathematical model (apparently for gold prices) you want it to optimise based on historic data. You can then use your model with those estimated parameters to predict the future cost.

If you have an example using ga to predict anything in any other context other than a fitted model, please share it.

[Shamir Sheikkariem]

I am using historical data (gold prices, shovel capacity, truck capacity etc) to predict the future cost (capital expenditure) of mining equipment (shovels and trucks).

[Star Strider]

If you have a mathematical espression that is a function of data and parameters that relates those data to the gold price, the ga function will provide the best estimate of the parameters for that expression.

If you have the data and the model function, I can likely help you estimate the parameters with the ga function.