Here I show how to produce P-value, S-value, likelihood, and deviance functions with the concurve
package using fake data and data from real studies. Simply put, these functions are rich sources of information for scientific inference and the image below, taken from Xie & Singh, 20131 displays why.
For a more extensive discussion of these concepts, see the following references.1–13
To get started, we could generate some normal data and combine two vectors in a dataframe
library(concurve)
set.seed(1031)
GroupA <- rnorm(500)
GroupB <- rnorm(500)
RandomData <- data.frame(GroupA, GroupB)
and look at the differences between the two vectors. We’ll plug these vectors and the dataframe they’re in inside of the curve_mean()
function. Here, the default method involves calculating CIs using the Wald method.
intervalsdf <- curve_mean(GroupA, GroupB,
data = RandomData, method = "default"
)
Each of the functions within concurve
will generally produce a list with three items, and the first will usually contain the function of interest.
tibble::tibble(intervalsdf[[1]])
#> # A tibble: 10,000 x 1
#> `intervalsdf[[1… $upper.limit $intrvl.width $intrvl.level $cdf $pvalue
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 -0.113 -0.113 0 0 0.5 1
#> 2 -0.113 -0.113 0.0000154 0.0001 0.500 1.00
#> 3 -0.113 -0.113 0.0000309 0.0002 0.500 1.00
#> 4 -0.113 -0.113 0.0000463 0.000300 0.500 1.00
#> 5 -0.113 -0.113 0.0000617 0.0004 0.500 1.00
#> 6 -0.113 -0.113 0.0000772 0.0005 0.500 1.00
#> 7 -0.113 -0.113 0.0000926 0.000600 0.500 0.999
#> 8 -0.113 -0.113 0.000108 0.0007 0.500 0.999
#> 9 -0.113 -0.112 0.000123 0.0008 0.500 0.999
#> 10 -0.113 -0.112 0.000139 0.0009 0.500 0.999
#> # … with 9,990 more rows, and 1 more variable: $svalue <dbl>
We can view the function using the ggcurve()
function. The two basic arguments that must be provided are the data argument and the “type” argument. To plot a consonance function, we would write “c
”.
(function1 <- ggcurve(data = intervalsdf[[1]], type = "c", nullvalue = TRUE))
We can see that the consonance “curve” is every interval estimate plotted, and provides the P-values, CIs, along with the median unbiased estimate It can be defined as such,
\[C V_{n}(\theta)=1-2\left|H_{n}(\theta)-0.5\right|=2 \min \left\{H_{n}(\theta), 1-H_{n}(\theta)\right\}\]
Its information counterpart, the surprisal function, can be constructed by taking the \(-log_{2}\) of the P-value.3,14,15
To view the surprisal function, we simply change the type to “s
”.
(function1 <- ggcurve(data = intervalsdf[[1]], type = "s"))
We can also view the consonance distribution by changing the type to “cdf
”, which is a cumulative probability distribution. The point at which the curve reaches 50% is known as the “median unbiased estimate”. It is the same estimate that is typically at the peak of the P-value curve from above.
(function1s <- ggcurve(data = intervalsdf[[2]], type = "cdf", nullvalue = TRUE))
We can also get relevant statistics that show the range of values by using the curve_table()
function. There are several formats that can be exported such as .docx, .ppt, and TeX.
(x <- curve_table(data = intervalsdf[[1]], format = "image"))
Lower Limit |
Upper Limit |
Interval Width |
Interval Level (%) |
CDF |
P-value |
S-value (bits) |
-0.132 |
-0.093 |
0.039 |
25.0 |
0.625 |
0.750 |
0.415 |
-0.154 |
-0.071 |
0.083 |
50.0 |
0.750 |
0.500 |
1.000 |
-0.183 |
-0.042 |
0.142 |
75.0 |
0.875 |
0.250 |
2.000 |
-0.192 |
-0.034 |
0.158 |
80.0 |
0.900 |
0.200 |
2.322 |
-0.201 |
-0.024 |
0.177 |
85.0 |
0.925 |
0.150 |
2.737 |
-0.214 |
-0.011 |
0.203 |
90.0 |
0.950 |
0.100 |
3.322 |
-0.233 |
0.008 |
0.242 |
95.0 |
0.975 |
0.050 |
4.322 |
-0.251 |
0.026 |
0.276 |
97.5 |
0.988 |
0.025 |
5.322 |
-0.271 |
0.046 |
0.318 |
99.0 |
0.995 |
0.010 |
6.644 |
If we wanted to compare two studies to see the amount of “consonance”, we could use the curve_compare()
function to get a numerical output.
First, we generate some more fake data
GroupA2 <- rnorm(500)
GroupB2 <- rnorm(500)
RandomData2 <- data.frame(GroupA2, GroupB2)
model <- lm(GroupA2 ~ GroupB2, data = RandomData2)
randomframe <- curve_gen(model, "GroupB2")
Once again, we’ll plot this data with ggcurve()
. We can also indicate whether we want certain interval estimates to be plotted in the function with the “levels
” argument. If we wanted to plot the 50%, 75%, and 95% intervals, we’d provide the argument this way:
(function2 <- ggcurve(type = "c", randomframe[[1]], levels = c(0.50, 0.75, 0.95), nullvalue = TRUE))
Now that we have two datasets and two functions, we can compare them using the curve_compare()
function.
(curve_compare(
data1 = intervalsdf[[1]], data2 = randomframe[[1]], type = "c",
plot = TRUE, measure = "default", nullvalue = TRUE
))
#> [1] "AUC = Area Under the Curve"
#> [[1]]
#>
#>
#> AUC 1 AUC 2 Shared AUC AUC Overlap (%) Overlap:Non-Overlap AUC Ratio
#> ------ ------ ----------- ---------------- ------------------------------
#> 0.098 0.073 0.024 16.309 0.195
#>
#> [[2]]
This function will provide us with the area that is shared between the curve, along with a ratio of overlap to non-overlap.
We can also do this for the surprisal function simply by changing type
to “s
”.
(curve_compare(
data1 = intervalsdf[[1]], data2 = randomframe[[1]], type = "s",
plot = TRUE, measure = "default", nullvalue = FALSE
))
#> [1] "AUC = Area Under the Curve"
#> [[1]]
#>
#>
#> AUC 1 AUC 2 Shared AUC AUC Overlap (%) Overlap:Non-Overlap AUC Ratio
#> ------ ------ ----------- ---------------- ------------------------------
#> 3.947 1.531 1.531 38.801 0.634
#>
#> [[2]]
It’s clear that the outputs have changed and indicate far more overlap than before.
Here, we’ll look at how to create consonance functions from the coefficients of predictors of interest in a Cox regression model.
We’ll use the carData
package for this. Fox & Weisberg, 2018 describe the dataset elegantly in their paper,
The Rossi data set in the
carData
package contains data from an experimental study of recidivism of 432 male prisoners, who were observed for a year after being released from prison (Rossi et al., 1980). The following variables are included in the data; the variable names are those used by Allison (1995), from whom this example and variable descriptions are adapted:week: week of first arrest after release, or censoring time.
arrest: the event indicator, equal to 1 for those arrested during the period of the study and 0 for those who were not arrested.
fin: a factor, with levels “yes” if the individual received financial aid after release from prison, and “no” if he did not; financial aid was a randomly assigned factor manipulated by the researchers.
age: in years at the time of release.
race: a factor with levels “black” and “other”.
wexp: a factor with levels “yes” if the individual had full-time work experience prior to incarceration and “no” if he did not.
mar: a factor with levels “married” if the individual was married at the time of release and “not married” if he was not.
paro: a factor coded “yes” if the individual was released on parole and “no” if he was not.
prio: number of prior convictions.
educ: education, a categorical variable coded numerically, with codes 2 (grade 6 or less), 3 (grades 6 through 9), 4 (grades 10 and 11), 5 (grade 12), or 6 (some post-secondary).
emp1–emp52: factors coded “yes” if the individual was employed in the corresponding week of the study and “no” otherwise.
We read the data file into a data frame, and print the first few cases (omitting the variables emp1 – emp52, which are in columns 11–62 of the data frame):
library(carData)
Rossi[1:5, 1:10]
#> week arrest fin age race wexp mar paro prio educ
#> 1 20 1 no 27 black no not married yes 3 3
#> 2 17 1 no 18 black no not married yes 8 4
#> 3 25 1 no 19 other yes not married yes 13 3
#> 4 52 0 yes 23 black yes married yes 1 5
#> 5 52 0 no 19 other yes not married yes 3 3
Thus, for example, the first individual was arrested in week 20 of the study, while the fourth individual was never rearrested, and hence has a censoring time of 52. Following Allison, a Cox regression of time to rearrest on the time-constant covariates is specified as follows:
library(survival)
mod.allison <- coxph(Surv(week, arrest) ~
fin + age + race + wexp + mar + paro + prio,
data = Rossi
)
mod.allison
#> Call:
#> coxph(formula = Surv(week, arrest) ~ fin + age + race + wexp +
#> mar + paro + prio, data = Rossi)
#>
#> coef exp(coef) se(coef) z p
#> finyes -0.37942 0.68426 0.19138 -1.983 0.04742
#> age -0.05744 0.94418 0.02200 -2.611 0.00903
#> raceother -0.31390 0.73059 0.30799 -1.019 0.30812
#> wexpyes -0.14980 0.86088 0.21222 -0.706 0.48029
#> marnot married 0.43370 1.54296 0.38187 1.136 0.25606
#> paroyes -0.08487 0.91863 0.19576 -0.434 0.66461
#> prio 0.09150 1.09581 0.02865 3.194 0.00140
#>
#> Likelihood ratio test=33.27 on 7 df, p=2.362e-05
#> n= 432, number of events= 114
Now that we have our Cox model object, we can use the curve_surv()
function to create the function.
If we wanted to create a function for the coefficient of prior convictions, then we’d do so like this:
z <- curve_surv(mod.allison, "prio")
Then we could plot our consonance curve and density and also produce a table of relevant statistics. Because we’re working with ratios, we’ll set the measure
argument in ggcurve()
to “ratio
”.
ggcurve(z[[1]], measure = "ratio", nullvalue = TRUE)
ggcurve(z[[2]], type = "cd", measure = "ratio", nullvalue = TRUE)
curve_table(z[[1]], format = "image")
Lower Limit |
Upper Limit |
Interval Width |
Interval Level (%) |
CDF |
P-value |
S-value (bits) |
1.086 |
1.106 |
0.020 |
25.0 |
0.625 |
0.750 |
0.415 |
1.075 |
1.117 |
0.042 |
50.0 |
0.750 |
0.500 |
1.000 |
1.060 |
1.133 |
0.072 |
75.0 |
0.875 |
0.250 |
2.000 |
1.056 |
1.137 |
0.080 |
80.0 |
0.900 |
0.200 |
2.322 |
1.052 |
1.142 |
0.090 |
85.0 |
0.925 |
0.150 |
2.737 |
1.045 |
1.149 |
0.103 |
90.0 |
0.950 |
0.100 |
3.322 |
1.036 |
1.159 |
0.123 |
95.0 |
0.975 |
0.050 |
4.322 |
1.028 |
1.168 |
0.141 |
97.5 |
0.988 |
0.025 |
5.322 |
1.018 |
1.180 |
0.162 |
99.0 |
0.995 |
0.010 |
6.644 |
We could also construct a function for another predictor such as age
x <- curve_surv(mod.allison, "age")
ggcurve(x[[1]], measure = "ratio")
ggcurve(x[[2]], type = "cd", measure = "ratio")
curve_table(x[[1]], format = "image")
Lower Limit |
Upper Limit |
Interval Width |
Interval Level (%) |
CDF |
P-value |
S-value (bits) |
0.938 |
0.951 |
0.013 |
25.0 |
0.625 |
0.750 |
0.415 |
0.930 |
0.958 |
0.028 |
50.0 |
0.750 |
0.500 |
1.000 |
0.921 |
0.968 |
0.048 |
75.0 |
0.875 |
0.250 |
2.000 |
0.918 |
0.971 |
0.053 |
80.0 |
0.900 |
0.200 |
2.322 |
0.915 |
0.975 |
0.060 |
85.0 |
0.925 |
0.150 |
2.737 |
0.911 |
0.979 |
0.068 |
90.0 |
0.950 |
0.100 |
3.322 |
0.904 |
0.986 |
0.081 |
95.0 |
0.975 |
0.050 |
4.322 |
0.899 |
0.992 |
0.093 |
97.5 |
0.988 |
0.025 |
5.322 |
0.892 |
0.999 |
0.107 |
99.0 |
0.995 |
0.010 |
6.644 |
That’s a very quick look at creating functions from Cox regression models.
Here, we’ll use an example dataset taken from the metafor
website, which also comes preloaded with the metafor
package.
library(metafor)
#> Loading required package: Matrix
#> Loading 'metafor' package (version 2.1-0). For an overview
#> and introduction to the package please type: help(metafor).
dat.hine1989
#> study source n1i n2i ai ci
#> 1 1 Chopra et al. 39 43 2 1
#> 2 2 Mogensen 44 44 4 4
#> 3 3 Pitt et al. 107 110 6 4
#> 4 4 Darby et al. 103 100 7 5
#> 5 5 Bennett et al. 110 106 7 3
#> 6 6 O'Brien et al. 154 146 11 4
I will quote Wolfgang here, since he explains it best,
"As described under help(dat.hine1989), variables n1i and n2i are the number of patients in the lidocaine and control group, respectively, and ai and ci are the corresponding number of deaths in the two groups. Since these are 2×2 table data, a variety of different outcome measures could be used for the meta-analysis, including the risk difference, the risk ratio (relative risk), and the odds ratio (see Table III). Normand (1999) uses risk differences for the meta-analysis, so we will proceed accordingly. We can calculate the risk differences and corresponding sampling variances with:
dat <- escalc(measure = "RD", n1i = n1i, n2i = n2i, ai = ai, ci = ci, data = dat.hine1989)
dat
#> study source n1i n2i ai ci yi vi
#> 1 1 Chopra et al. 39 43 2 1 0.0280 0.0018
#> 2 2 Mogensen 44 44 4 4 0.0000 0.0038
#> 3 3 Pitt et al. 107 110 6 4 0.0197 0.0008
#> 4 4 Darby et al. 103 100 7 5 0.0180 0.0011
#> 5 5 Bennett et al. 110 106 7 3 0.0353 0.0008
#> 6 6 O'Brien et al. 154 146 11 4 0.0440 0.0006
"Note that the yi values are the risk differences in terms of proportions. Since Normand (1999) provides the results in terms of percentages, we can make the results directly comparable by multiplying the risk differences by 100 (and the sampling variances by \(100^{2}\)):
We can fit a fixed-effects model with the following
fe <- rma(yi, vi, data = dat, method = "FE")
Now that we have our metafor object, we can compute the consonance function using the curve_meta()
function.
fecurve <- curve_meta(fe)
Now we can graph our function.
ggcurve(fecurve[[1]], nullvalue = TRUE)
We used a fixed-effects model here, but if we wanted to use a random-effects model, we could do so with the following, which will use a restricted maximum likelihood estimator for the random-effects model
re <- rma(yi, vi, data = dat, method = "REML")
And then we could use curve_meta()
to get the relevant list
recurve <- curve_meta(re)
Now we can plot our object.
ggcurve(recurve[[1]], nullvalue = TRUE)
We could also compare our two models to see how much consonance/overlap there is
curve_compare(fecurve[[1]], recurve[[1]], plot = TRUE)
#> [1] "AUC = Area Under the Curve"
#> [[1]]
#>
#>
#> AUC 1 AUC 2 Shared AUC AUC Overlap (%) Overlap:Non-Overlap AUC Ratio
#> ------ ------ ----------- ---------------- ------------------------------
#> 2.085 2.085 2.085 100 Inf
#>
#> [[2]]
The results are practically the same and we cannot actually see any difference, and the AUC % overlap also indicates this.
We can also take a set of confidence limits and use them to construct a consonance, surprisal, likelihood or deviance function using the curve_rev()
function. This method is computed from the approximate normal distribution.
Here, we’ll use two epidemiological studies16,17 that studied the impact of SSRI exposure in pregnant mothers, and the rate of autism in children.
Both of these studies suggested a null effect of SSRI exposure on autism rates in children.
curve1 <- curve_rev(point = 1.7, LL = 1.1, UL = 2.6, type = "c", measure = "ratio", steps = 10000)
(ggcurve(data = curve1[[1]], type = "c", measure = "ratio", nullvalue = TRUE))
curve2 <- curve_rev(point = 1.61, LL = 0.997, UL = 2.59, type = "c", measure = "ratio", steps = 10000)
(ggcurve(data = curve2[[1]], type = "c", measure = "ratio", nullvalue = TRUE))
The null value is shown via the red line and it’s clear that a large mass of the function is away from it.
We can also see this by plotting the likelihood functions via the curve_rev()
function.
lik1 <- curve_rev(point = 1.7, LL = 1.1, UL = 2.6, type = "l", measure = "ratio", steps = 10000)
(ggcurve(data = lik1[[1]], type = "l1", measure = "ratio", nullvalue = TRUE))
lik2 <- curve_rev(point = 1.61, LL = 0.997, UL = 2.59, type = "l", measure = "ratio", steps = 10000)
(ggcurve(data = lik2[[1]], type = "l1", measure = "ratio", nullvalue = TRUE))
We can also view the amount of agreement between the likelihood functions of these two studies.
(plot_compare(
data1 = lik1[[1]], data2 = lik2[[1]], type = "l1", measure = "ratio", nullvalue = TRUE, title = "Brown et al. 2017. J Clin Psychiatry. vs. \nBrown et al. 2017. JAMA.",
subtitle = "J Clin Psychiatry: OR = 1.7, 1/6.83 LI: LL = 1.1, UL = 2.6 \nJAMA: HR = 1.61, 1/6.83 LI: LL = 0.997, UL = 2.59", xaxis = expression(Theta ~ "= Hazard Ratio / Odds Ratio")
))
and the consonance functions
(plot_compare(
data1 = curve1[[1]], data2 = curve2[[1]], type = "c", measure = "ratio", nullvalue = TRUE, title = "Brown et al. 2017. J Clin Psychiatry. vs. \nBrown et al. 2017. JAMA.",
subtitle = "J Clin Psychiatry: OR = 1.7, 1/6.83 LI: LL = 1.1, UL = 2.6 \nJAMA: HR = 1.61, 1/6.83 LI: LL = 0.997, UL = 2.59", xaxis = expression(Theta ~ "= Hazard Ratio / Odds Ratio")
))
Some authors have shown that the bootstrap distribution is equal to the confidence distribution because it meets the definition of a consonance distribution.1,18,19 The bootstrap distribution and the asymptotic consonance distribution would be defined as:
\[H_{n}(\theta)=1-P\left(\hat{\theta}-\hat{\theta}^{*} \leq \hat{\theta}-\theta | \mathbf{x}\right)=P\left(\hat{\theta}^{*} \leq \theta | \mathbf{x}\right)\]
Certain bootstrap methods such as the BCa
method and t
-bootstrap method also yield second order accuracy of consonance distributions.
\[H_{n}(\theta)=1-P\left(\frac{\hat{\theta}^{*}-\hat{\theta}}{\widehat{S E}^{*}\left(\hat{\theta}^{*}\right)} \leq \frac{\hat{\theta}-\theta}{\widehat{S E}(\hat{\theta})} | \mathbf{x}\right)\]
Here, I demonstrate how to use these particular bootstrap methods to arrive at consonance curves and densities.
We’ll use the Iris
dataset and construct a function that’ll yield a parameter of interest.
iris <- datasets::iris
foo <- function(data, indices) {
dt <- data[indices, ]
c(
cor(dt[, 1], dt[, 2], method = "p")
)
}
We can now use the curve_boot()
method to construct a function. The default method used for this function is the “Bca
” method provided by the bcaboot
package.19
I will suppress the output of the function because it is unnecessarily long. But we’ve placed all the estimates into a list object called y.
The first item in the list will be the consonance distribution constructed by typical means, while the third item will be the bootstrap approximation to the consonance distribution.
ggcurve(data = y[[1]], nullvalue = TRUE)
ggcurve(data = y[[3]], nullvalue = TRUE)
We can also print out a table for TeX documents
(gg <- curve_table(data = y[[1]], format = "image"))
Lower Limit |
Upper Limit |
Interval Width |
Interval Level (%) |
CDF |
P-value |
S-value (bits) |
-0.142 |
-0.093 |
0.048 |
25 |
0.625 |
0.75 |
0.415 |
-0.169 |
-0.067 |
0.102 |
50 |
0.750 |
0.50 |
1.000 |
-0.205 |
-0.031 |
0.174 |
75 |
0.875 |
0.25 |
2.000 |
-0.214 |
-0.021 |
0.194 |
80 |
0.900 |
0.20 |
2.322 |
-0.266 |
0.031 |
0.296 |
95 |
0.975 |
0.05 |
4.322 |
-0.312 |
0.077 |
0.389 |
99 |
0.995 |
0.01 |
6.644 |
More bootstrap replications will lead to a smoother function. But for now, we can compare these two functions to see how similar they are.
plot_compare(y[[1]], y[[3]])
If we wanted to look at the bootstrap standard errors, we could do so by loading the fifth item in the list
knitr::kable(y[[5]])
theta | sdboot | z0 | a | sdjack | |
---|---|---|---|---|---|
est | -0.1175698 | 0.0755961 | 0.0576844 | 0.0304863 | 0.075694 |
jsd | 0.0000000 | 0.0010234 | 0.0274023 | 0.0000000 | 0.000000 |
where in the top row, theta
is the point estimate, and sdboot
is the bootstrap estimate of the standard error, sdjack
is the jacknife estimate of the standard error. z0
is the bias correction value and a
is the acceleration constant.
The values in the second row are essentially the internal standard errors of the estimates in the top row.
The densities can also be calculated accurately using the t
-bootstrap method. Here we use a different dataset to show this
library(Lock5Data)
dataz <- data(CommuteAtlanta)
func <- function(data, index) {
x <- as.numeric(unlist(data[1]))
y <- as.numeric(unlist(data[2]))
return(mean(x[index]) - mean(y[index]))
}
Our function is a simple mean difference. This time, we’ll set the method to “t
” for the t
-bootstrap method
z <- curve_boot(data = CommuteAtlanta, func = func, method = "t", replicates = 2000, steps = 1000)
#> Warning in norm.inter(t, alpha): extreme order statistics used as endpoints
ggcurve(data = z[[1]], nullvalue = FALSE)
ggcurve(data = z[[2]], type = "cd", nullvalue = FALSE)
The consonance curve and density are nearly identical. With more bootstrap replications, they are very likely to converge.
(zz <- curve_table(data = z[[1]], format = "image"))
Lower Limit |
Upper Limit |
Interval Width |
Interval Level (%) |
CDF |
P-value |
S-value (bits) |
-39.400 |
-39.075 |
0.325 |
25.0 |
0.625 |
0.750 |
0.415 |
-39.611 |
-38.876 |
0.735 |
50.0 |
0.750 |
0.500 |
1.000 |
-39.873 |
-38.608 |
1.265 |
75.0 |
0.875 |
0.250 |
2.000 |
-39.932 |
-38.530 |
1.402 |
80.0 |
0.900 |
0.200 |
2.322 |
-40.026 |
-38.456 |
1.570 |
85.0 |
0.925 |
0.150 |
2.737 |
-40.118 |
-38.354 |
1.763 |
90.0 |
0.950 |
0.100 |
3.322 |
-40.294 |
-38.174 |
2.120 |
95.0 |
0.975 |
0.050 |
4.322 |
-40.442 |
-38.026 |
2.416 |
97.5 |
0.988 |
0.025 |
5.322 |
-40.636 |
-37.806 |
2.830 |
99.0 |
0.995 |
0.010 |
6.644 |
For the examples above, we mainly used nonparametric bootstrap methods. Here I show an example using the parametric Bca
bootstrap and the results it yields.
First, we’ll load our data again and set our function.
data(diabetes, package = "bcaboot")
X <- diabetes$x
y <- scale(diabetes$y, center = TRUE, scale = FALSE)
lm.model <- lm(y ~ X - 1)
mu.hat <- lm.model$fitted.values
sigma.hat <- stats::sd(lm.model$residuals)
t0 <- summary(lm.model)$adj.r.squared
y.star <- sapply(mu.hat, rnorm, n = 1000, sd = sigma.hat)
tt <- apply(y.star, 1, function(y) summary(lm(y ~ X - 1))$adj.r.squared)
b.star <- y.star %*% X
Now, we’ll use the same function, but set the method to “bcapar
” for the parametric method.
df <- curve_boot(method = "bcapar", t0 = t0, tt = tt, bb = b.star)
Now we can look at our outputs.
ggcurve(df[[1]], nullvalue = FALSE)
ggcurve(df[[3]], nullvalue = FALSE)
We can compare the functions to see how well the bootstrap approximations match up
plot_compare(df[[1]], df[[3]])
We can also look at the density function
ggcurve(df[[5]], type = "cd", nullvalue = FALSE)
That concludes our demonstration of the bootstrap method to approximate consonance functions.
For this last example, we’ll explore the curve_lik()
function, which can help generate profile likelihood functions, and deviance statistics with the help of the ProfileLikelihood
package.
library(ProfileLikelihood)
#> Loading required package: nlme
#> Loading required package: MASS
We’ll use a simple example taken directly from the ProfileLikelihood
documentation where we’ll calculate the likelihoods from a glm model
data(dataglm)
xx <- profilelike.glm(y ~ x1 + x2,
data = dataglm, profile.theta = "group",
family = binomial(link = "logit"), length = 500, round = 2
)
#> Warning message: provide lo.theta and hi.theta
Then, we’ll use curve_lik()
on the object that the ProfileLikelihood
package created.
lik <- curve_lik(xx, dataglm)
tibble::tibble(lik[[1]])
#> # A tibble: 500 x 1
#> `lik[[1]]`$values $likelihood $loglikelihood $support $deviancestat
#> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 -1.41 9.26e-21 -9.79 0.0000560 9.79
#> 2 -1.40 1.00e-20 -9.71 0.0000606 9.71
#> 3 -1.39 1.08e-20 -9.63 0.0000655 9.63
#> 4 -1.38 1.17e-20 -9.56 0.0000708 9.56
#> 5 -1.37 1.26e-20 -9.48 0.0000765 9.48
#> 6 -1.35 1.37e-20 -9.40 0.0000826 9.40
#> 7 -1.34 1.47e-20 -9.32 0.0000892 9.32
#> 8 -1.33 1.59e-20 -9.25 0.0000963 9.25
#> 9 -1.32 1.72e-20 -9.17 0.000104 9.17
#> 10 -1.31 1.85e-20 -9.10 0.000112 9.10
#> # … with 490 more rows
Next, we’ll plot three functions, the relative likelihood, the log-likelihood, the likelihood, and the deviance function.
ggcurve(lik[[1]], type = "l1", nullvalue = TRUE)
ggcurve(lik[[1]], type = "l2")
ggcurve(lik[[1]], type = "l3")
ggcurve(lik[[1]], type = "d")
The obvious advantage of using reduced likelihoods is that they are free of nuisance parameters
\[L_{t_{n}}(\theta)=f_{n}\left(F_{n}^{-1}\left(H_{p i v}(\theta)\right)\right)\left|\frac{\partial}{\partial t} \psi\left(t_{n}, \theta\right)\right|=h_{p i v}(\theta)\left|\frac{\partial}{\partial t} \psi(t, \theta)\right| /\left.\left|\frac{\partial}{\partial \theta} \psi(t, \theta)\right|\right|_{t=t_{n}}\] thus, giving summaries of the data that can be incorporated into combined analyses.
1. Xie M-g, Singh K. Confidence Distribution, the Frequentist Distribution Estimator of a Parameter: A Review. International Statistical Review. 2013;81(1):3-39. doi:10.1111/insr.12000
2. Birnbaum A. A unified theory of estimation, I. The Annals of Mathematical Statistics. 1961;32(1):112-135. doi:10.1214/aoms/1177705145
3. Chow ZR, Greenland S. Semantic and Cognitive Tools to Aid Statistical Inference: Replace Confidence and Significance by Compatibility and Surprise. arXiv:190908579 [statME]. September 2019. http://arxiv.org/abs/1909.08579.
4. Fraser DAS. P-Values: The Insight to Modern Statistical Inference. Annual Review of Statistics and Its Application. 2017;4(1):1-14. doi:10.1146/annurev-statistics-060116-054139
5. Fraser DAS. The P-value function and statistical inference. The American Statistician. 2019;73(sup1):135-147. doi:10.1080/00031305.2018.1556735
6. Poole C. Beyond the confidence interval. American Journal of Public Health. 1987;77(2):195-199. doi:10.2105/AJPH.77.2.195
7. Poole C. Confidence intervals exclude nothing. American Journal of Public Health. 1987;77(4):492-493. doi:10.2105/ajph.77.4.492
8. Schweder T, Hjort NL. Confidence and Likelihood*. Scand J Stat. 2002;29(2):309-332. doi:10.1111/1467-9469.00285
9. Schweder T, Hjort NL. Confidence, Likelihood, Probability: Statistical Inference with Confidence Distributions. Cambridge University Press; 2016.
10. Singh K, Xie M, Strawderman WE. Confidence distribution (CD) – distribution estimator of a parameter. August 2007. http://arxiv.org/abs/0708.0976.
11. Sullivan KM, Foster DA. Use of the confidence interval function. Epidemiology. 1990;1(1):39-42. doi:10.1097/00001648-199001000-00009
12. Whitehead J. The case for frequentism in clinical trials. Statistics in Medicine. 1993;12(15-16):1405-1413. doi:10.1002/sim.4780121506
13. Rothman KJ, Greenland S, Lash TL. Precision and statistics in epidemiologic studies. In: Rothman KJ, Greenland S, Lash TL, eds. Modern Epidemiology. 3rd ed. Lippincott Williams & Wilkins; 2008:148-167.
14. Greenland S. Valid P-values behave exactly as they should: Some misleading criticisms of P-values and their resolution with S-values. The American Statistician. 2019;73(sup1):106-114. doi:10.1080/00031305.2018.1529625
15. Shannon CE. A mathematical theory of communication. The Bell System Technical Journal. 1948;27(3):379-423. doi:10.1002/j.1538-7305.1948.tb01338.x
16. Brown HK, Ray JG, Wilton AS, Lunsky Y, Gomes T, Vigod SN. Association between serotonergic antidepressant use during pregnancy and autism spectrum disorder in children. JAMA. 2017;317(15):1544-1552. doi:10.1001/jama.2017.3415
17. Brown HK, Hussain-Shamsy N, Lunsky Y, Dennis C-LE, Vigod SN. The association between antenatal exposure to selective serotonin reuptake inhibitors and autism: A systematic review and meta-analysis. The Journal of Clinical Psychiatry. 2017;78(1):e48-e58. doi:10.4088/JCP.15r10194
18. Efron B, Tibshirani RJ. An Introduction to the Bootstrap. CRC Press; 1994.
19. Efron B, Narasimhan B. The automatic construction of bootstrap confidence intervals. October 2018:17.