Background The yin deficiency type of perimenopausal syndrome (PMS) as a common category of PMS based on the theory of traditional Chinese medicine (TCM) has a high prevalence with severe symptoms and long course of disease. Therefore, it is necessary to construct a prediction model to assist in diagnosis. Objective This study aimed to investigate the independent predictors of the yin deficiency type of PMS and to develop a clinical prediction model of this disease. Methods PMS patients who attended the Third Affiliated Hospital of Zhejiang Chinese Medical University between February 2020 and August 2023 were selected and divided chronologically into training and validation groups. Logistic regression analysis was applied in the training group to clarify the independent predictors of the yin deficiency type of PMS, and a nomogram was plotted. Internal and external validations were performed in the training and validation groups to evaluate the model's accuracy, goodness of fit, and clinical adaptability. Results Hot flashes and sweating (≥10 episodes/day), palpitations, emotional fluctuations, and abnormal sexual activity were independent predictors of the yin deficiency type of PMS (P > 0.05). Based on the clinical prediction model constructed, the area under the receiver operating characteristic curve (AUR OC) in the training group was 0.989 (95%CI 0.980–0.998), and the AUR OC in the validation group was 0.971 (95%CI 0.940–0.999). This demonstrates that the model has superior prediction performance. The Hosmer-Lemeshow test was used to evaluate the model's goodness of fit with P = 0.596 for the training group and P = 0.883 for the validation group, indicating a good fit. The decision curve analysis (DCA) curve and clinical impact curve (CIC) indicated good clinical adaptability. Conclusion The model can accurately predict the occurrence of the yin deficiency type of PMS, which may help clinicians identify such patients at an early stage.