import java.io.IOException;

import com.ampl.AMPL;
import com.ampl.Objective;
import com.ampl.Parameter;
import com.ampl.Variable;
import com.ampl.Environment;

public class EfficientFrontier {

  public static void main(String[] args) throws IOException {
    // Number of steps of the efficient frontier
    int steps = 10;
    // Use the provided path or the default one
    String baseDir = args.length > 1 ? args[1] : "../models";
    String modelDirectory = baseDir + "/qpmv";

    // Create an AMPL instance
    AMPL ampl = new AMPL();

    /*
     * // If the AMPL installation directory is not in the system search path:
     * Environment env = new Environment(
     * "full path to the AMPL installation directory");
     * AMPL ampl = new AMPL(env);
     */

    // Outer try-catch-finally block, to be sure of releasing the AMPL
    // object when done
    try {
      if (args.length > 0)
        if (!args[0].equals("NA"))
          ampl.setOption("solver", args[0]);

      ampl.setBoolOption("reset_initial_guesses", true);
      ampl.setBoolOption("send_statuses", false);

      // Load the AMPL model from file
      ampl.read(modelDirectory + "/qpmv.mod");
      ampl.read(modelDirectory + "/qpmvbit.run");

      // Set tables directory (parameter used in the script above)
      ampl.getParameter("data_dir").set(modelDirectory);
      // Read tables
      ampl.readTable("assetstable");
      ampl.readTable("astrets");

      Variable portfolioReturn = ampl.getVariable("portret");
      Parameter averageReturn = ampl.getParameter("averret");
      Parameter targetReturn = ampl.getParameter("targetret");
      Objective deviation = ampl.getObjective("cst");

      ampl.eval("let stockopall:={};let stockrun:=stockall;");

      // Relax the integrality
      ampl.setBoolOption("relax_integrality", true);
      // Solve the problem
      ampl.solve();
      // Calibrate the efficient frontier range
      double minret = portfolioReturn.value();
      double maxret = findMax(averageReturn.getValues().getColumnAsDoubles("averret"));
      double stepsize = (maxret - minret) / steps;

      double[] returns = new double[steps];
      double[] deviations = new double[steps];

      for (int i = 0; i < steps; i++) {
        System.out.format("Solving for return = %f%n", maxret - (i - 1) * stepsize);
        // Set target return to the desired point
        targetReturn.setValues(maxret - (i - 1) * stepsize);
        ampl.eval("let stockopall:={};let stockrun:=stockall;");
        // Relax integrality
        ampl.setBoolOption("relax_integrality", true);
        ampl.solve();
        System.out.format("QP result = %f %n", deviation.value());
        // Adjust included stocks
        ampl.eval("let stockrun:={i in stockrun:weights[i]>0};");
        ampl.eval("let stockopall:={i in stockrun:weights[i]>0.5};");
        // Set integrality back
        ampl.setBoolOption("relax_integrality", false);
        ampl.solve();
        System.out.format("QMIP result = %f%n", deviation.value());
        // Store data of corrent frontier point
        returns[i] = maxret - (i - 1) * stepsize;
        deviations[i] = deviation.value();
      }

      // Display efficient frontier points
      System.out.format("%-8s  %-8s%n", "RETURN", "DEVIATION");
      for (int i = 0; i < returns.length; i++)
        System.out.format("%-6f  %-6f%n", returns[i], deviations[i]);
    } finally {
      ampl.close();
    }
  }

  private static double findMax(double[] array) {
    double max = Double.NEGATIVE_INFINITY;
    for (double d : array)
      if (d > max)
        max = d;
    return max;
  }
}