["Journal of Child Psychology and Psychiatry, EarlyView. ", "\n\nBackground\nChild internalizing and externalizing behavioral problems are highly prevalent psychiatric symptoms worldwide, for which maternal prenatal stress is a known risk factor. However, underlying neuroendocrine mechanisms remain largely unclear. We investigated whether maternal hair cortisol (HCC) and cortisone concentration (HCNC) are associated with offspring's internalizing and externalizing behavior problems in a prospective pre‐birth cohort study from Perú.\n\n\nMethods\nN = 271 mother–child dyads were included in this analysis. Recruitment and data collection took place at the Instituto Nacional Materno Perinatal in Lima, Perú. HCC and HCHC were obtained from hair segments representing up to 3 months pre‐pregnancy and first trimester, respectively, and were quantified via liquid chromatography tandem mass spectrometry. The Child Behavior Checklist was used to assess internalizing and externalizing behavioral problems of children (mean age at follow‐up = 6.98 years (SD = 1.05)). Marginal structural models estimated population average associations between HCC, HCNC, and internalizing and externalizing problems, adjusting for established covariates.\n\n\nResults\nAt pre‐pregnancy, logHCNC was positively associated with offspring internalizing (β = 2.21, 95% CI: 0.46; 3.96, p = .013) and externalizing problems (β = 1.87, 95% CI: 0.34; 3.40, p = .016). At the first trimester, logHCNC was negatively associated with internalizing (β = −2.51, 95%CI: −4.37; −0.64, p = .008), and externalizing problems (β = −2.73, 95% CI: −4.18; −1.28, p < .001). Associations were stronger for females and not apparent for logHCC.\n\n\nConclusions\nWe found time‐dependent associations between stress‐related prenatal hair glucocorticoid concentration and offspring behavioral problems. Modeling biomarker data time‐dependently may prove critical to identifying the underlying mechanisms of transgenerational stress transmission.\n\n"]