formatting {CALIBERdatamanage} | R Documentation |
Functions to format numbers for publication, producing either plain text or LaTeX output. formatnum
is the general function. formatp
formats P values. formathr
exponentiates and prints coefficients such as hazard ratios with confidence intervals, and displays significance levels with stars.
formatnum(numbers, dp = NA, sigfig = 3, scientific = TRUE, latex = (getOption("xtable.type") == "latex"), usephantom = getOption("CALIBERdatamanage_phantom"), names = NULL, decimal.mark = ".") formatp(pvalues, minp = 0.0001, latex=(getOption('xtable.type')=='latex'), ...) formathr(coef, se, df = Inf, dp = NA, pstar = TRUE, pvalueci = 0.05, template = '\\1 (\\2, \\3)', ...) formatci(range, scientific = FALSE, template = '(\\1, \\2)', ...)
numbers |
numeric vector of numbers to format |
dp |
number of decimal places |
sigfig |
number of significant figures |
scientific |
TRUE for scientific mode, FALSE for standard output |
latex |
TRUE for LaTeX output (math mode), FALSE, NULL or logical(0) for plain text output |
usephantom |
TRUE to pad the LaTeX output with blank (phantom) characters so that the decimal points line up |
names |
names for the output vector |
pvalues |
numeric vector of P value values to be formatted |
minp |
minimum P value to display as an exact value |
coef |
coefficient (log hazard ratio or log odds ratio) |
se |
standard error of coefficient |
df |
degrees of freedom for t-distribution; Inf to use the Normal distribution |
pstar |
TRUE to include stars for P value: * < 0.05, ** < 0.01, *** < 0.001 |
range |
numeric vector containing the upper and lower confidence limits |
pvalueci |
significance level to be used in the calculation of confidence intervals |
template |
character string template with \1 and \2 representing numbers forming the confidence interval (formatci), and \1, \2 and \3 representing the estimate and confidence interval for formathr. |
decimal.mark |
character to be used for decimal point, default is . |
... |
other arguments to be passed to |
a character vector containing formatted numbers
Anoop Shah
percent
, npercent
, percentConf
, nmissing
, meansd
formatnum(1:5, latex = FALSE) # [1] "1" "2" "3" "4" "5" formatnum(1:5, latex = TRUE) # [1] "$1$" "$2$" "$3$" "$4$" "$5$" formatnum(c(0.00001, 0.02, 5.6, 67, 700), latex = FALSE, dp = 1) # [1] "0.0" "0.0" "5.6" "67.0" "700.0" formatnum(c(0.00001, 0.02, 5.6, 67, 700), latex = TRUE, usephantom = TRUE) # [1] "$1\times 10^{-5}$" # [2] "\phantom{0}\phantom{0}$0.02$" # [3] "\phantom{0}\phantom{0}$5.6\phantom{0}$" # [4] "\phantom{0}$67\phantom{.}\phantom{0}\phantom{0}$" # [5] "$700\phantom{.}\phantom{0}\phantom{0}$" # Formatting hazard ratios formathr(0.8, 0.3, latex = FALSE) # "2.23 (1.24, 4.01) **" formathr(0.8, 0.3, latex = TRUE) # "$2.23$ ($1.24$, $4.01$) **" formathr(0.8, 0.3, latex = TRUE, template = '\1; \1--\2') # "$2.23$; $1.24$--$4.01$ **" # Missing values should not cause errors formathr(c(0.8, 0.8, NA), c(NA, 0.3, 0.3)) # [1] "2.23 (NA, NA)" "2.23 (1.24, 4.01) **" "NA (NA, NA)" # Formatting confidence intervals formatci(c(2.4343, 4.3245), latex = FALSE) # "(2.43, 4.32)" formatci(c(2.4343, 4.3245), latex = TRUE) # "($2.43$, $4.32$)"