The disabilities affecting the peripheral regions of the body can be often as a result of cerebellar damage. Tremor, poor and inaccurate coordination, and irregular movements associated with gait, balance and speech are some of the manifestations. Conventionally expert opinion determines severity of Cerebellar Ataxia (CA) and the assessment is likely to be inherently subjective. The automated versions of two commonly used tests: Finger to Nose test (FNT) and Heel to Shin Test (HST), are investigated in this paper for evaluating disability and movement deficits due to CA. Limb movements are measured using Inertial Measurement Units (IMU) that captures the disability related information, using kinematic parameters such as acceleration and angular velocity considered in both time and frequency domain. Using the Singular Spectrum Analysis (SSA), the collective dominance in the data distributions of the underlying features were observed. The dominant features were combined to substantiate the correlation with the expert clinical assessments through Linear Discriminant and Regression analysis where the classifier performance was also verified by means of cross-validation. This study found that IMU features captured characteristic movements as intention tremor in FNT and not in HST. In FNT predominantly consisting of translational movements, the rotation was a dominant feature whereas acceleration were observed to be more dominant for the case of HST.