Skip to main content

Table 2 Selection of input variables1, 2

From: Predicting the growth performance of growing-finishing pigs based on net energy and digestible lysine intake using multiple regression and artificial neural networks models

Step 1

Step 23

Variables

P-value

Selection

Forms3

Selection

BW

<0.01

√

Linear

√

NE intake

<0.01

√

Quadratic

√

CP intake

0.30

 

Cubic

 

SID Lys intake

<0.01

√

  

SID Met intake

0.29

   

SID Thr intake

0.10

   

SID Trp intake

0.37

   

SID Val intake

0.11

   

ADF intake

0.29

   

NDF intake

0.21

   
  1. ADF acid detergent fiber, ADFI average daily feed intake, ADG average daily gain, BW body weight, CP crude protein, F/G feed conversion ratio, NDF neutral detergent fiber, NE net energy, SID standardized ileal digestible
  2. 1 Step 1 was used to select the most sensitive variables to predict ADG and F/G, and Step 2 aimed to find the appropriate forms of input variables.
  3. 2 No significant interactive effects of the selected variables in Step 1 were detected, with P-value > 0.05.
  4. 3 The R2 for predicting ADG in different forms were: linear: 0.89; quadratic: 0.93; cubic: 0.93. The R2 for predicting G/F in different forms were: linear: 0.87; quadratic: 0.88; cubic: 0.88.
Back to article page

Journal of Animal Science and Biotechnology

ISSN: 2049-1891