This study aims to perform age estimation using three different parameters from dental and skeletal development. The sample consisted of 98 dental records of patients aged from 10 to 16 years old, containing the chronological age and a set of radiographs (panoramic, lateral cephalometric and carpal radiographs) taken in the same day. The biological age was assessed through the dental development from panoramic radiographs according to the Nicodemo's method. The stages of dental development were registered and imported in CRONOL software (UNESP, São Paulo, Brazil), which provided the estimated dental age. The lateral cephalometric radiographs were analyzed to assess the development of the vertebrae C2, C3 and C4. And carpal radiographs were evaluated according to Fishman's method. Shapiro-Wilk test was used to verify the normality of the chronological and estimated age. T-test for unpaired samples was used to compare the normal data. Chi-square test was used to analyze the age in function of sex. Moderate and strong correlations were found between the chronological and biological (estimated) ages for all the methods. Statistically significant differences between the development of males and females were not observed (p>0.05). A linear regression formula was designed to allow age estimates statistically more accurate (p<0.0001). The formula reached an accuracy rate of 71.4%. In general, the methods underestimated the age. The combination of methods led to optimal performances for age estimation. The performances were similar between males and females.