Objective To establish and validate a nomogram model for predicting the risk of loneliness among elderly people living alone in China. Methods Data from the 2018 China Longitudinal Health and Longevity Survey (CLHLS) were used to select 2,252 elderly people aged 65 years and older living alone, then they were randomly assigned in a 7∶3 ratio to a training group (1,576 cases) and a validation group (676 cases). Lasso regression was conduct to screen variables, multivariate logistic regression analysis was performed to identify factors influencing loneliness among elderly people living alone, a predictive nomogram model was established, and its discriminatory ability and calibration were assessed by using ROC curves and the Hosmer-Lemeshow test. External validation was conducted among 306 elderly people living alone in Linghe District, Jinzhou City, Liaoning Province. Results Among the 2,252 elderly people living alone, 1,138 (50.5%) experienced loneliness. Results indicated that, self-rated health status, sleep quality, community care, source of livelihood, hearing difficulties, and loss of interest were included in the predictive model (all P<0.05). The area under the ROC curve for the training group and the validation group was 0.769 (95%CI: 0.746-0.792) and 0.740 (95% CI:0.704-0.777), respectively and the Hosmer-Lemeshow goodness-of-fit test showed good model fit (training group P=0.167, validation group P=0.071, external validation P=0.351). Conclusion The risk of loneliness among elderly people living alone in China is high, and the nomogram model demonstrates good predictive performance, making it a reliable tool for predicting loneliness among elderly people living alone.