Table 7 The 8SC for models before removing heterogeneity parameters.

From: Hybrid models of sparse and robust regression to solve heterogeneity problem in black pepper big data

Machine learning models

Highest Important Variables

AIC

FPE

GCV

HQ

RICE

SCHWARZ

SGMASQ

SHIBATA

Elastic Net

25

64.43607

64.43618

64.44795

66.24278

64.46004

69.46604

63.57703

64.41296

35

58.35419

58.35444

58.37488

60.63182

58.39611

64.75461

57.28263

58.31433

45

54.83302

54.83352

54.86488

60.29664

54.89782

62.63196

53.55314

54.77229

55

54.15528

54.16052

54.31018

57.58240

54.47740

62.51839

52.45918

53.87690

100

53.94150

53.94238

53.98812

57.25186

54.03668

63.42084

51.41674

53.85357

Ridge

25

100.0249

100.0250

100.0433

102.8294

100.0621

107.83293

98.69136

99.98898

35

62.76986

62.77013

62.79212

65.21984

62.81496

69.6546

61.61721

62.72699

45

60.55568

60.55623

60.59086

63.59199

60.62724

69.16855

59.14223

60.48862

55

59.24027

59.24125

59.29148

62.87582

59.34481

69.65079

57.56574

59.14371

100

57.12154

57.12706

57.28492

63.59928

57.46130

76.49045

54.27776

56.82792

LASSO

25

64.91212

64.91223

64.92409

66.73218

64.93627

69.97925

64.04673

64.88884

35

61.09099

61.09126

61.11265

63.47544

61.13488

67.79159

59.96917

61.04927

45

55.33281

55.33331

55.36496

58.10725

55.39820

63.20283

54.04127

55.27153

55

54.98971

54.99062

55.03724

58.36440

55.08675

64.65326

53.43533

54.90008

100

54.33572

54.34096

54.49113

60.49753

54.65890

72.76000

51.63063

54.05641