Skip to main content
. 2020 Jul 14;34(2):306–311. doi: 10.5713/ajas.20.0293

Table 4.

Prediction equations for energy concentrations in feed ingredients and validation of equations1),2)

Equation No. n Equation3) RMSE R2 Intercept4) Slope5)
Value p-value Value p-value
1 364 DE = 398+0.79×GE−44.90×ADF 355 0.66 27 (70) 0.701 −0.29 (0.13) 0.028
2 371 DE = 728+0.76×GE−25.18×NDF 363 0.64 −49 (63) 0.442 −0.22 (0.12) 0.066
3 343 DE = 536+0.84×GE−42.33×ash−23.87×NDF 330 0.71 −33 (62) 0.590 −0.25 (0.11) 0.025
4 344 DE = 3,558+14.49×CP+52.06×EE−54.68×ADF 354 0.66 43 (72) 0.555 −0.44 (0.11) <0.001
5 353 DE = 3,901−40.07×ash+14.71×CP−40.79×ADF 411 0.55 −5 (75) 0.952 −0.40 (0.13) 0.004
6 351 DE = 3,882+8.78×CP+55.38×EE−29.05×NDF 397 0.56 −56 (65) 0.391 −0.30 (0.11) 0.008
7 320 DE = 3,665−57.10×ash+17.86×CP+50.57×EE−47.72×ADF 328 0.71 84 (68) 0.225 −0.40 (0.10) 0.001
8 329 DE = 3,969−80.49×ash+14.37×CP+66.42×EE−25.19×NDF 325 0.71 −1 (59) 0.983 −0.30 (0.09) 0.002
9 425 ME = 0.97×DE−3.86×CP 120 1.00 43 (10) <0.001 0.06 (0.02) 0.001
10 369 ME = 568+0.71×GE−43.30×ADF 379 0.58 73 (78) 0.355 −0.32 (0.15) 0.038
11 372 ME = 965+0.66×GE−24.62×NDF 370 0.60 −23 (73) 0.754 −0.26 (0.14) 0.072
12 345 ME = 1,133+0.65×GE−29.05×ash−23.17×NDF 338 0.67 −26 (69) 0.713 −0.21 (0.13) 0.112
13 342 ME = 781+0.69×GE−30.88×ash−40.87×ADF 347 0.66 66 (75) 0.384 −0.29 (0.14) 0.044
14 331 ME = 4,178−48.19×ash+60.43×EE−28.79×NDF 355 0.63 −23 (64) 0.721 −0.26 (0.11) 0.018
15 361 ME = 3,854−56.40×ash+11.12×CP−38.36×ADF 399 0.53 36 (80) 0.655 −0.41 (0.14) 0.007
16 328 ME = 3,662−72.14×ash+13.73×CP+46.32×EE−46.24×ADF 327 0.69 120 (75) 0.117 −0.41 (0.12) 0.001
17 331 ME = 3,903−64.52×ash+8.70×CP+64.49×EE−26.43×NDF 321 0.70 9 (66) 0.897 −0.28 (0.11) 0.014

RMSE, root mean square of error; DE, digestible energy; GE, gross energy; ADF, acid detergent fiber; NDF, neutral detergent fiber; CP, crude protein; EE, ether extract; ME, metabolizable energy.

1)

Thirty nine in vivo data points were employed to validate prediction equations for energy concentrations in feed ingredients.

2)

Values in parentheses are standard error.

3)

Energy concentrations and chemical composition are expressed as kcal/kg dry matter and % dry matter, respectively.

4)

The intercept represents the mean bias.

5)

The slope represents the linear bias.