Genetic markers are widely used for selection and prediction in commercial pig populations. Number of piglets born alive (NBA) and stillborn piglets (STB) within litters have a major effect on overall economic returns. The aim of this study was to determine the association between estrogen receptor gene (ESR) and paternally expressed gene 1 (PEG1) and NBA and STB. The data set included information from 2 commercial pig populations, one in Northern Thailand (210 sows; 123 Landrace and 87 Yorkshire) and another one in Northeastern Thailand (130 sows; 70 Landrace and 60 Large White). Each population was analyzed separately using 2 models. Model 1 included farrowing year-season, parity, age at first farrowing and breed group as fixed effects, and animal and residual as random effects. Model 2 included all effects from model 1 plus genotypes for ESR (AA, AB, and BB) and PEG1 (AA, GA, and GG) as fixed effects. Association between predicted EBV from models 1 and 2 were analyzed using Spearman rank correlations. Genotype frequencies were 0.67, 0.29 and 0.04 for ESR and 0.01, 0.40, and 0.59 for PEG1 in the Northern population, and 0.42, 0.34, and 0.24 for ESR and 0.05, 0.25, and 0.70 for PEG1 in the Northeastern population. The ESR genotype influenced the EBV of Yorkshire sows for NBA (P < 0.05) and STB (P < 0.01). Yorkshire ESR AA sows had the highest average EBV for NBA (0.16 ± 0.03 piglets), whereas Yorkshire ESR AB sows had the lowest average EBV for STB (−0.01 ± 0.01 piglets). The PEG1 genotype affected the EBV of Landrace, Yorkshire and Large White sows for NBA and STB (P < 0.05). Large White PEG1 GG sows had the highest average EBV for NBA (0.43 ± 0.06 piglets), and Landrace PEG1 AA sows had the lowest average EBV for STB (−0.09 ± 0.09 piglets). Positive rank correlations between EBV from models 1 and 2 existed in the 2 populations for all breeds (0.41 to 0.63; P < 0.01), except for Landrace sows in the Northern population (−0.08; P > 0.05). These rank correlations indicated that selecting sows with higher NBA and lower STB by using EBV from models with and without ESR and PEG1 marker effects would likely yield similar outcomes.