Accuracy Analysis of Financial Distress Prediction Models for Companies on the IDX Watchlist Board in 2020-2022 Period

Abstract

Abstrak: Tujuan penelitian adalah menguji tingkat keakuratan model prediksi financial distress Altman, Grover, Zmijewski, dan Springate, menggunakan data keuangan perusahaan yang masuk kriteria 5 atau 8 di Papan Pemantauan Khusus BEI pada periode 2020- 2022. Penelitian ini menggunakan metode kuantitatif dengan teknis analisis deskriptif. Pengujian data menggunakan uji Kruskal Wallis One Way Anova karena terdapat lebih dari dua model prediksi yang dibandingkan, kemudian data tidak berdistribusi normal. Sebanyak 44 sampel yang dipilih menggunakan teknik puposive diambil dari 55 populasi emiten yang masuk dalam Papan Pemantauan Khusus BEI. Hasil penelitian menunjukkan bahwa tidak terdapat perbedaan yang signifikan antara akurasi model Altman, Grover, Zmijewski, dan Springate. Ditunjukkan dari hasil akurasi berdasarkan jumlah prediksi benar dari masing-masing model, Zmijewski merupakan model paling akurat dengan tingkat akurasi 67%, disusul Altman 65%, Grover 65%, dan terendah Springate 60%
 
 Abstract: The research aims to examine the accuracy level of the Altman, Grover, Zmijewski, and Springate financial distress prediction models, to determine the most accurate financial distress prediction model in analyzing companies on the IDX Watchlist Board criteria 5 or 8 between 2020 and 2022. This study employs a quantitative method with descriptive analytical techniques. Data testing utilizes the Kruskal-Wallis One-Way ANOVA test due to the comparison of more than two prediction models and the non-normal distribution of the data. A total of 44 samples, purposively selected from the population of 55 listed issuers within the DPK-BEI, were used. The research findings reveal no significant differences in the accuracy among the Altman, Grover, Zmijewski, and Springate models. This is evidenced by the accuracy results based on the number of correct predictions from each model. Zmijewski emerges as the most accurate model with a 67% accuracy rate, followed by Altman and Grover at 65% each, and Springate with the lowest accuracy at 60%.