CQERDDF

class pystoned.CQERDDF.CERDDF(y, x, b=None, gy=[1], gx=[1], gb=None, fun='prod', tau=0.5)

Convex expectile regression with DDF formulation

__init__(y, x, b=None, gy=[1], gx=[1], gb=None, fun='prod', tau=0.5)

CER DDF

Parameters:

• y (float) – output variable.

• x (float) – input variables.

• b (float), optional) – undesirable output variables. Defaults to None.

• gy (list, optional) – output directional vector. Defaults to [1].

• gx (list, optional) – input directional vector. Defaults to [1].

• gb (list, optional) – undesirable output directional vector. Defaults to None.

• fun (String, optional) – FUN_PROD (production frontier) or FUN_COST (cost frontier). Defaults to FUN_PROD.

• tau (float, optional) – expectile. Defaults to 0.5.

display_alpha()

Display alpha value

display_beta()

Display beta value

display_delta()

Display delta value

display_gamma()

Display gamma value

display_lamda()

Display lamda value

display_negative_residual()

Dispaly negative residual value

display_positive_residual()

Dispaly positive residual value

display_residual()

Dispaly residual value

display_status()

Display the status of problem

get_adjusted_alpha()

Return the shifted constatnt(alpha) term by CCNLS

get_adjusted_residual()

Return the shifted residuals(epsilon) tern by CCNLS

get_alpha()

Return alpha value by array

get_beta()

Return beta value by array

get_delta()

Return delta value by array

get_frontier()

Return estimated frontier value by array

get_gamma()

Return gamma value by array

get_lamda()

Return lamda value by array

get_negative_residual()

Return negative residual value by array

get_positive_residual()

Return positive residual value by array

get_predict(x_test)

Return the estimated function in testing sample

get_residual()

Return residual value by array

get_status()

Return status

optimize(email='local', solver=None)

Optimize the function by requested method

Parameters:

• email (string) – The email address for remote optimization. It will optimize locally if OPT_LOCAL is given.

• solver (string) – The solver chosen for optimization. It will optimize with default solver if OPT_DEFAULT is given.

class pystoned.CQERDDF.CQRDDF(y, x, b=None, gy=[1], gx=[1], gb=None, fun='prod', tau=0.5)

Convex quantile regression with directional distance function

__init__(y, x, b=None, gy=[1], gx=[1], gb=None, fun='prod', tau=0.5)

CQR DDF

Parameters:

• y (float) – output variable.

• x (float) – input variables.

• b (float), optional) – undesirable output variables. Defaults to None.

• gy (list, optional) – output directional vector. Defaults to [1].

• gx (list, optional) – input directional vector. Defaults to [1].

• gb (list, optional) – undesirable output directional vector. Defaults to None.

• fun (String, optional) – FUN_PROD (production frontier) or FUN_COST (cost frontier). Defaults to FUN_PROD.

• tau (float, optional) – quantile. Defaults to 0.5.

display_alpha()

Display alpha value

display_beta()

Display beta value

display_delta()

Display delta value

display_gamma()

Display gamma value

display_lamda()

Display lamda value

display_negative_residual()

Dispaly negative residual value

display_positive_residual()

Dispaly positive residual value

display_residual()

Dispaly residual value

display_status()

Display the status of problem

get_adjusted_alpha()

Return the shifted constatnt(alpha) term by CCNLS

get_adjusted_residual()

Return the shifted residuals(epsilon) tern by CCNLS

get_alpha()

Return alpha value by array

get_beta()

Return beta value by array

get_delta()

Return delta value by array

get_frontier()

Return estimated frontier value by array

get_gamma()

Return gamma value by array

get_lamda()

Return lamda value by array

get_negative_residual()

Return negative residual value by array

get_positive_residual()

Return positive residual value by array

get_predict(x_test)

Return the estimated function in testing sample

get_residual()

Return residual value by array

get_status()

Return status

optimize(email='local', solver=None)

Optimize the function by requested method

Parameters:

• email (string) – The email address for remote optimization. It will optimize locally if OPT_LOCAL is given.

• solver (string) – The solver chosen for optimization. It will optimize with default solver if OPT_DEFAULT is given.