A new heavy-tailed distribution called the mixture Pareto-loggamma distribution has been developed by combining the classical Pareto and a special case of the loggamma distribution. This model is useful for analyzing insurance data. The researchers used log-moments and maximum likelihood to estimate the parameters of the distribution. They also found that composite models can be easily created by matching the mode of the distribution. The performance of the new distribution and composite models was tested using different claims datasets.