It is known that fault modeling and fault propagation in analog circuits are extremely important and more challenging than in digital circuits. Several automated model generation (AMG) techniques are developed to model the nonlinear behavior of faulty analog circuits. However, most of the modeling techniques are performed under the MATLAB environment which is impractical and the models cannot be utilized in electronic circuits. To perform high level fault modeling (HLFM) and fault propagation (FP) on system level, the models need to be translated into hardware description language (HDL) models such as VHDL-AMS or Verilog-AMS models. In this paper, several faults are modeled for transistor level analog circuits using nonlinear autoregressive exogenous (NLARX) AMG technique in MATLAB. The resulting MATLAB models are translated into VHDL-AMS behavioral models. HLFM and FP are successfully implemented for benchmark analog circuits: inverting amplifier and biquadratic low-pass filter circuits. © 2011 IEEE.