Thank you. How do I then handle the model statement? ... View more

Thanks for your response. I want to get the best A, B, C values to model IVGP. using a non linear regression model. Can you help me please.? Data ACC1; Input Time SPECIES$ IVGP; Cards; 2     C1    10.49293 2     C1    8.328766 2     C11   5.771113 2     C11   4.787401 2     CN2   6.164598 2     CN2   5.377628 2     CN21  4.787401 2     CN21  3.606946 2     CN3   1.836263 2     CN3   1.04687 2     CN31  3.881162 2     CN31  3.150836 2     U2    8.328766 2     U2    9.641 2     U21   2.426491 2     U21   2.623233 2     U3    9.312478 2     U3    7.148311 2     U31   4.197173 2     U31   6.558083 2     U1    9.509221 2     U1    11.87013 2     U11   7.854721 2     U11   9.014867 2     CN1   12.26362 2     CN1   10.29619 2     CN11  5.565581 2     CN11  5.328766 4     C1    15.72311 4     C1    13.19804 4     C11   9.837126 4     C11   8.263186 4     CN2   8.853413 4     CN2   6.885988 4     CN21  5.705533 4     CN21  6.689246 4     CN3   3.48732 4     CN3   2.34672 4     CN31  6.840228 4     CN31  6.696743 4     U2    10.6241 4     U2    11.30363 4     U21   5.705533 4     U21   5.902276 4     U3    13.57523 4     U3    12.39478 4     U31   7.082731 4     U31   11.41107 4     U1    16.34592 4     U1    18.08402 4     U11   13.65219 4     U11   14.99125 4     CN1   19.08402 4     CN1   17.1166 4     CN11  10.78808 4     CN11  8.165461 8     C1    24.95372 8     C1    22.06774 8     C11   15.32963 8     C11   14.77198 8     CN2   15.54266 8     CN2   12.00129 8     CN21  9.066443 8     CN21  11.00129 8     CN3   8.129712 8     CN3   7.697125 8     CN31  9.918563 8     CN31  8.754396 8     U2    18.90357 8     U2    20.30325 8     U21   11.80455 8     U21   12.59152 8     U3    19.87099 8     U3    19.47751 8     U31   14.16546 8     U31   18.4938 8     U1    23.05145 8     U1    24.3635 8     U11   19.87432 8     U11   20.14679 8     CN1   24.19933 8     CN1   24.00259 8     CN11  13.45993 8     CN11  14.28077 12    C1    36.59454 12    C1    35.74112 12    C11   25.05188 12    C11   24.65839 12    CN2   26.62582 12    CN2   20.72354 12    CN21  17.64081 12    CN21  17.51051 12    CN3   19.73983 12    CN3   15.01801 12    CN31  13.24733 12    CN31  10.68968 12    U2    32.7737 12    U2    33.14078 12    U21   23.87142 12    U21   24.26491 12    U3    29.97044 12    U3    31.34764 12    U31   27.41279 12    U31   32.33135 12    U1    35.72484 12    U1    36.41408 12    U11   29.05188 12    U11   32.2662 12    CN1   36.85643 12    CN1   35.67597 12    CN11  24.78869 12    CN11  23.34764 24    C1    75.56455 24    C1    73.50069 24    C11   60.85901 24    C11   60.07204 24    CN2   58.69485 24    CN2   50.23492 24    CN21  48.44666 24    CN21  49.51439 24    CN3   50.82515 24    CN3   46.10333 24    CN31  45.28378 24    CN31  43.97173 24    U2    61.61212 24    U2    61.19814 24    U21   64.59712 24    U21   64.00689 24    U3    64.94174 24    U3    63.46553 24    U31   70.69614 24    U31   72.86031 24    U1    72.33523 24    U1    72.59841 24    U11   64.28507 24    U11   66.28637 24    CN1   74.63099 24    CN1   72.86031 24    CN11  58.72871 24    CN11  58.26749 48    C1    144.8848 48    C1    143.2118 48    C11   117.7669 48    C11   115.832 48    CN2   90.09264 48    CN2   90.58256 48    CN21  88.82816 48    CN21  90.66658 48    CN3   84.20536 48    CN3   90.96105 48    CN31  119.1278 48    CN31  118.9962 48    U2    92.84445 48    U2    90.86418 48    U21   149.163 48    U21   151.3284 48    U3    131.282 48    U3    130.7127 48    U31   130.178 48    U31   129.9812 48    U1    121.5213 48    U1    125.6529 48    U11   139.2144 48    U11   141.7995 48    CN1   145.0177 48    CN1   143.0789 48    CN11  122.1128 48    CN11  120.9298 ; PROC SORT DATA = WORK.ACC1; BY SPECIES; RUN; proc sgplot data=WORK.ACC; scatter y = IVGP x=TIME / group=species; loess y =IVGP x=time / group=species;  /* estimate curves */ run; PROC NLIN BEST = 9; BY SPECIES; PARMS A = 9 TO 15 BY 1       B = 39 TO 44 BY 1         C = 0.03 TO 0.05 BY 0.005; MODEL IVGP = A + B*(1-EXP(-C*TIME)); OUTPUT OUT = NlinOut PARMS = A B C; RUN; ... View more

Sir, Thanks for your response and assistance. Howeevr, my effort to use the new syntax you gave me showed nice curves of the gas production but the output values looks disjointed or maybe i don seem to be able to interprete it. I was hoping to get results showing ,e "a" "b" and "C" values which will help me calculate Efective gas produced using the model.  ... View more

Thank you for your response. My treatments were different feed stuffs designated as species and codenamed as : C1, C11, CN1, CN11, CN2, CN21, Cn3, CN31, U1,U11,U2,U21,U3,U31 (14 treatments). Data was collected as a cummulative methane gas production across times 2 h, 4h,8h,12h,24 and 48Hours after incubation of species. I want to model the effective gas produced as well as the a, b, c values for each species. I do hope my answer relates o what you asked?    ... View more

Dear Steven, Your my objective aligns with your thoughts on the analysis. However, while trying to practice with 2 of the parameters (24H Gas and 24H Methane), i kept getting an Error Mesaage stating as follows:              200 ERROR 22-322: Syntax error, expecting one of the following: (, /, =. ERROR 200-322: The symbol is not recognized and will be ignored. The SAS syntax and actual data input was: Title 'Gas and methane'; Data Gasandmethane; Label v1= Gas24, v2=Methane24; input Treatment$ Replicate v1-v2; cards; Control 1 74.1129 15.91843325 Control 2 71.3585 17.9904604 Control 3 62.11161 15.71941815 Urea1 1 75.4901 15.04878412 Urea1 2 63.68555 11.51948683 Urea1 3 69.98131 16.30495324 Urea2 1 56.01259 10.69626777 Urea2 2 60.34092 11.40405239 Urea2 3 70.76828 17.17851915 Urea3 1 77.26078 17.3410505 Urea3 2 65.45623 14.48968449 Urea3 3 79.62169 11.58406475 Nitrate1 1 59.35721 7.178511302 Nitrate1 2 67.42365 14.80763354 Nitrate1 3 56.79956 10.86226791 Nitrate2 1 45.58524 5.513145744 Nitrate2 2 56.60282 8.706394519 Nitrate2 3 52.86471 7.230013325 Nitrate3 1 48.73312 2.611616701 Nitrate3 2 46.76569 3.318005431 Nitrate3 3 56.60282 5.138831879 ; Proc Glimmix Data = Gasandmethane; Class Replicate Treatment; Model v1-v2 = Treatment;/ Random intercept / Subject = Replicate;/ title 'Gas and Methane from Urea and Nitrate-Orthogonal contrasts'; /*N-Source main effect*/ Contrast 'Source' Treatment 0  1 1  1  -1  -1  -1; /*N-Level main effect*/ Contrast 'Level' Treatment     0 +1  -1  0  +1   -1  0, Treatment     0  +1 0  -1  +1 0  -1; /*Interaction effect*/ Contrast 'Source x Level'  Treatment     0 +1  -1   0  -1   +1 0,                                               Treatment     0 +1  0   -1 -1    0  +1; /*Control vs all N-supplemented*/ Contrast 'Control Vs. All' Treatment     +6 -1  -1  -1 -1  -1  -1;         /*Orthogonal polynomials for trend comparison of Level effect (averaged over frequencies, control included)*/ Contrast 'Linear Effect for Level' Treatment  -0.67082 -0.1118035  0.1118035  0.3354102 -0.1118035  0.1118035  0.3354102;; Contrast 'Quadratic Effect for Level' Treatment 0.5  -0.25  -0.25 0.25  -0.25  -0.25 0.25; Contrast 'Cubic Effect for Level' Treatment  -0.223607 0.3354102  -0.3354102 0.1118035  0.3354102 -0.3354102  0.1118035;*/ LSMeans Treatment /DIFF lines;   Run; I shall be grateful for your guidance. Thank You. Festus ... View more

Dear Steven, Thank you very much. I will apply Proc GLIMMIX as you advised. The 21 variables were not weights taken at successive periods. They include Gas production at 24 hours, Methane production, Feed digestibility, Ammonia NItrogen production etc...All the parameters are derived from the animals following the digestion of the experimental diets. You may also help me to clarify if i will be in order to 1. Evaluate an interaction effect of N-Source and Level using Level 1 Vs Leve 2; and Level 1 Vs Level 3 (what about Level 2 Vs 3) as shown in the SAS command i prepared      /*Interaction effect*/          CONTRAST 'Source x Level'  TRMT 0     +1        -1         0          -1         +1        0,                                                       TRMT 0      +1        0          -1         -1         0          +1;    2. If i have to do the trend comparison (linear, quadratic and cubi effect of N-level) separately from the  previous contrasts because the number of orthogonal contrast alowed is just 6 based on my 2X3 factorial +1. ... View more

Dear Steven, I am a beginer with the use of SAS and particulalarly Orthogonal contrast. Your contributions above has really helped me create a contrast command for my analysis and wish to seek your opinion on it. Will this be effective to interprete the results of my data??? My experiment involve the use of 2 sources of Nitrogen at 3 levels of inclusion plus a control as diet additives to feed experimental animals.The response was simulated in the laboratory and the procedure repeated 3 times to give 3 replicates. The various responses are labeled as variables 1-21 as depicted in the SAS command i just prepared. I used Proc IML to get the coefficients for the polynomial effects of N-level. Here is what i have: N-sources= 2 N-Levels =3 Control=1 Total treatments= 7 PROC GLM; CLASS Treatment Level Replicate; MODEL v1-v21= Replicate Treatment Level Treatment*Level; /*Control vs all N-supplemented*/ CONTRAST 'CTRL Vs. All' Treatment +6  -1  -1  -1  -1  -1  -1; /*N-Source main effect*/ CONTRAST 'Urea Vs. Nitrate' Treatment 0  1  1  1  -1  -1  -1; /*N-Level main effect*/ CONTRAST 'Level' TRMT 0        +1        -1         0          +1        -1         0,                              TRMT 0        +1        0          -1         +1        0          -1; /*Interaction effect*/ CONTRAST 'Source x Level'  TRMT 0     +1        -1         0          -1         +1        0,                                            TRMT 0      +1        0          -1         -1         0          +1;         /*Orthogonal polynomials for trend comparison of Level effect (averaged over frequencies, control included)*/ CONTRAST 'Linear Rate Effect for Level' Treatment  -0.67082  -0.1118035  0.1118035  0.3354102  -0.1118035  0.1118035  0.3354102; CONTRAST 'Quadratic Rate Effect for Level' Treatment 0.5  -0.25  -0.25  0.25  -0.25  -0.25  0.25; CONTRAST 'Cubic Rate Effect for Level' Treatment  -0.223607  0.3354102  -0.3354102  0.1118035  0.3354102  -0.3354102  0.1118035;*/ LSMeans Treatment /DIFF lines;   Run; Thanks in anticipation of your kind suggestions and advise ... View more