The study aim was to improve the prediction of CH₄ emissions from beef cattle by developing equations specific for high forage diets (≥40% DM basis). Treatment means from 38 beef studies published between 2000 and 2014 with dietary forage ≥40% DM were compiled into a database. Criteria for inclusion in the database were intake, diet composition and enteric CH₄ production. Principal component analysis detected relevant variables associated with CH₄. Because of the limited size of the original database (n = 123), a Monte Carlo technique was used to resample 1,000 times the data from each study to create a new virtual data set. Outliers were excluded by Mahalanobis distance in both the original database and virtual data set. Using the original database, forward stepwise multiple regression was used to obtain prediction equations. The random effect of study was included in the analysis using the Mixed procedure and ‘leave-one-out’ cross validation was used to internally validate the equations. Using the virtual data set (n = 100,305), equations were developed using forward stepwise multiple regression and K-fold cross validation (n = 10). Model performance was evaluated as observed-predicted values using concordance correlation (r_c) and root mean square prediction error (RMSPE, g/d). Statistical analysis was performed using JMP v11. Using the original database, the best-fit equation was: CH₄ (g/d) = 71.5(±11.45) + 0.12(±0.03) × BW (kg) + 0.10(±0.01) × DMI³ (kg/d) − 244.8(±56.44) × fat³ (kg/d) with P < 0.0001, r_c: 0.72 and RMSPE: 39.1; where BW = body weight; DMI = dry matter intake. Using the Monte Carlo data set, the best-fit equation was: CH₄ (g/d) = 25.9(±0.54) + 0.13(±0.001) × BW (kg) + 145.4 (±1.31) × fat (kg/d) + 10.3(±0.16) × (NDF − ADF)² (kg/d) + 0.1(±0.00) × DMI³ (kg/d) − 27.4 (±0.20) × (starch/NDF) with P < 0.0001, r_c: 0.81 and RMSPE: 36.4 where NDF = neutral detergent fiber and ADF = acid detergent fiber. Monte Carlo data set equation improved prediction accuracy compared with the original database equation, but extensive feed analysis is required to use the equation. Both equations specifically developed for beef cattle fed forage diets may increase the accuracy of predicting CH₄ production.