The determinants of corporate bond credit spreads are investigated in Chile as an example of an emerging market with relatively few actors and thin trading. Both market-level and firm-level factors are considered. Three models previously used to analyze the highly developed US market are applied to Chilean inflation-indexed bond trade data, and the results for the two markets are compared. The determinants found to be significant for Chile form the basis for the design of a new multifactor regression model that is used to explain Chilean bond spreads. The results are evaluated with an out-of-sample test, and the root-mean-square error is calculated to compare the model's results with those obtained by the method commonly applied in illiquid markets by repeating the last recorded transaction for days on which no data are available. The proposed formulation is found to reduce the degree of error.