lambda_z estimation [🇷 for BE/BA]

posted by yjlee168 Homepage – Kaohsiung, Taiwan, 2008-09-28 02:08 (5682 d 14:07 ago) – Posting: # 2428
Views: 56,572

Dear Ace,

Thanks for your assistance. I look through your codes and find that your codes pick the correct numbers of data points this time. So I change one line of your codes from return(n_lambda) to summary(lm(log(conc)~time,dat[(nrow(dat)-n_lambda+1):nrow(dat),])) because I want to see if the final results are the same with those obtained from WinNonlin. Our original code for this line is summary(lm(log(conc)~time,dat[i:nrow(dat),])) which apparently does not include n_lambda in there. So the final results are the following (dumping from R console):

f<- function(dat) {
  n_lambda=0
  r.adj=0
  for (i in (nrow(dat)-2):(which.max(dat$conc+1))) {
  if (r.adj - summary(lm(log(conc)~time,dat[i:nrow(dat),]))$adj.r.squared <
     (0.0001)) {
     n_lambda <- nrow(dat)-i+1
     r.adj <- summary(lm(log(conc)~time,dat[i:nrow(dat),]))$adj.r.squared
  }
  }
  summary(lm(log(conc)~time,dat[(nrow(dat)-n_lambda+1):nrow(dat),]))
  }
b<-c(0,0.25,0.5,0.75,1,1.5,2,3,4,8,12,24)
<-c(0,36.1,125,567,963,1343,1739,1604,1460,797,383,72)
at <- data.frame(time=b,conc=c)
f(dat)

Call:
lm(formula = log(conc) ~ time, data = dat[(nrow(dat) - n_lambda +
    1):nrow(dat), ])

Residuals:
       8        9       10       11       12
-0.01329  0.04253  0.03672 -0.09658  0.03062

Coefficients:
             Estimate Std. Error t value Pr(>|t|)   
(Intercept)  7.843188   0.050394  155.64 5.85e-07 ***
time        -0.149881   0.003962  -37.83 4.06e-05 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.06733 on 3 degrees of freedom
Multiple R-squared: 0.9979,     Adjusted R-squared: 0.9972
F-statistic:  1431 on 1 and 3 DF,  p-value: 4.063e-05

b<-c(0,0.25,0.5,0.75,1,1.5,2,3,4,8,12,24)
d<-c(0,84.5,192,629,873,1246,1633,1375,1006,616,379,84.4)
dat <- data.frame(time=b,conc=d)
f(dat)

Call:
lm(formula = log(conc) ~ time, data = dat[(nrow(dat) - n_lambda +
    1):nrow(dat), ])

Residuals:
         9         10         11         12
-0.0057874 -0.0002764  0.0100142 -0.0039504

Coefficients:
              Estimate Std. Error t value Pr(>|t|)   
(Intercept)  7.4155262  0.0081685   907.8 1.21e-06 ***
time        -0.1240003  0.0005776  -214.7 2.17e-05 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.008645 on 2 degrees of freedom
Multiple R-squared:     1,      Adjusted R-squared: 0.9999
F-statistic: 4.609e+04 on 1 and 2 DF,  p-value: 2.170e-05

b<-c(0,0.5,0.75,1,1.5,2,3,4,8,12,24)
c<-c(0,69.7,167,602,1023,1388,1481,1346,658,336,84)
dat <- data.frame(time=b,conc=c)
f(dat)

Call:
lm(formula = log(conc) ~ time, data = dat[(nrow(dat) - n_lambda +
    1):nrow(dat), ])

Residuals:
       9       10       11
 0.07269 -0.09692  0.02423

Coefficients:
            Estimate Std. Error t value Pr(>|t|) 
(Intercept)  7.42148    0.16961   43.76   0.0145 *
time        -0.12562    0.01049  -11.97   0.0530 .
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.1236 on 1 degrees of freedom
Multiple R-squared: 0.9931,     Adjusted R-squared: 0.9861
F-statistic: 143.4 on 1 and 1 DF,  p-value: 0.05305

Bingo! The final results are exactly the same as those obtained from WinNonlin with these three examples. We will test more examples to confirm this.
—
All the best,
-- Yung-jin Lee
bear v2.9.1:- created by Hsin-ya Lee & Yung-jin Lee
Kaohsiung, Taiwan https://www.pkpd168.com/bear
Download link (updated) -> here
Thread locked

Complete thread:

UA Flag
Activity
 Admin contact
22,987 posts in 4,824 threads, 1,665 registered users;
85 visitors (0 registered, 85 guests [including 4 identified bots]).
Forum time: 16:15 CEST (Europe/Vienna)

The only way to comprehend what mathematicians mean by Infinity
is to contemplate the extent of human stupidity.    Voltaire

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5