Function reference • wisclabmisc

Analyses 🔬

fit_beta_gamlss() fit_beta_gamlss_se() predict_beta_gamlss() optimize_beta_gamlss_slope() uniroot_beta_gamlss(): Fit a beta regression model (for intelligibility)

fit_gen_gamma_gamlss() fit_gen_gamma_gamlss_se() predict_gen_gamma_gamlss(): Fit a generalized gamma regression model (for speaking rate)

GAMLSS helpers ⛑️

mem_gamlss(): Fit a gamlss model but store user data

check_sample_centiles(): Compute the percentage of points under each centile line

predict_centiles() pivot_centiles_longer(): Predict and tidy centiles from a GAMLSS model

ROC statistics 🥅

compute_smooth_density_roc(): Create an ROC curve from smoothed densities

compute_empirical_roc(): Create an ROC curve from observed data

compute_predictive_value_from_rates(): Compute positive and negative predictive value

trapezoid_auc() partial_trapezoid_auc(): Compute AUCs using the trapezoid method

Data cleaning 🧹

tocs_item() tocs_type() tocs_length(): Extract the TOCS details from a string (usually a filename)

Other functions 📌

format_year_month_age(): Convert age in months to years;months

chrono_age(): Compute chronological age in months

fit_kmeans(): Run (scaled) k-means on a dataset.

impute_values_by_length(): Staged imputation

join_to_split(): Join data onto resampled IDs

logitnorm_mean(): Compute the mean of logit-normal distribution(s)

tocs_item() tocs_type() tocs_length(): Extract the TOCS details from a string (usually a filename)

weight_lengths_with_ordinal_model(): Weight utterance lengths by using an ordinal regression model

Datasets 🗺️

data_example_intelligibility_by_length: Simulated intelligibility scores by utterance length

data_fake_intelligibility: Fake intelligibility data

data_fake_rates: Fake speaking rate data

data_features_consonants data_features_vowels: Phonetic features of consonants and vowels