The prediction of mechanical and thermal properties of 3D printed short-fiber reinforced polymers (SFRPs) are investigated in this study. Methods are demonstrated for predicting the internal spatially varying fiber orientation state and resulting internal spatially varying stiffness, coefficient of thermal expansion, and strength properties in a single bead of 13% carbon fiber filled acrylonitrile butadiene styrene. The methods allow determination of both the spatially varying microstructural properties and the effective, bulk properties in any direction by finite element analysis. The focus of this work is specifically on Large Area Additive Manufacturing, an extrusion-based process for manufacturing thermoplastic parts that are several feet long, but the methods are applicable to other SFRP processing methods as well. For the experimental validation portion of this dissertation, a large-scale 3D printing system was constructed to fabricate test specimens. Tensile, compressive, and flexural specimens were fabricated with this system and tested. It is demonstrated that correct order of magnitude predictions can be made for the effective stiffness, CTE, and strength of LAAM-printed SFRP beads using the presented computational methodology. In addition, the computational methodology lays a framework that lends itself to improvement by using more accurate modeling inputs as they are measured, and more accurate underlying equations as they are developed.