The Emissions Prediction and Policy Analysis (EPPA) model is a component of the MIT Integrated Earth Systems Model (IGSM). Here, we provide an overview of the model accessible to a broad audience and present the detailed structure, data, and parameterization of the model for specialists in economic modeling. EPPA projects emissions of most of the climatically important substances emitted as a result of human activities including carbon dioxide (CO2), carbon monoxide (CO), methane (CH4), nitrous oxide (N2O), nitrogen oxides (NOx), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), sulfur hexafluoride (SF6), sulfate aerosols (SOx), non-methane volatile organic compounds (NMVOCs), black carbon, organic carbon, and ammonia (NH3). We present an updated and consistent inventory for 1995 of all of these emissions disaggregated to the regional and sectoral levels we use in EPPA. This more complete inventory of climatically important substances shows non-energy sources (e.g. agriculture, biomass burning) and developing countries to be important current sources of many of these emissions. A major use of EPPA, a computable general equilibrium model of the world economy with regional and sectoral detail, is to estimate the cost of greenhouse gas emissions control over the 100-year horizon of the model. Reference projections show rates of improvement in energy use per unit of output (Gross Domestic Product) consistent with historical rates although in EPPA we do not attempt to model short-term business cycle behavior so that our projections do not show the same variability as the historical data. Emissions of climatically important substances mostly grow over time in our reference projection (although rates differ substantially among them) despite considerable improvements in energy efficiency and reductions in emissions coefficients for other substances. Developing countries as group become larger sources of all greenhouse gas emissions than developed or transition economies by the middle of the century as their economies and populations are projected to grow more rapidly. There remain many uncertainties in projections of this type. The projections presented in this report are a starting point (i.e. reference) for evaluating alternative scenarios and climate policies.