Not yet reviewed by Pith; the record is open.

We study the core shift effect in the parsec scale jet of the blazar 3C 454.3 using the 4.8 GHz - 36.8 GHz radio light curves obtained from three decades of continuous monitoring. From a piecewise Gaussian fit to each flare, time lags $\Delta t$ between the observation frequencies $\nu$ and spectral indices $\alpha$ based on peak amplitudes $A$ are determined. From the fit $\Delta t \propto \nu^{1/k_r}$, $k_r = 1.10 \pm 0.18$ indicating equipartition between the magnetic field energy density and the particle energy density. From the fit $A \propto \nu^\alpha$, $\alpha$ is in the range $-0.24$ to $1.52$. A mean magnetic field strength at 1 pc, $B_1 = 0.5 \pm 0.2$ G, and at the core, $B_{\rm core} = 46 \pm 16$ mG, are inferred, consistent with previous estimates. The measure of core position offset is $\Omega_{r\nu} = 6.4 \pm 2.8$ pc GHz$^{1/k_r}$ when averaged over all frequency pairs. Based on the statistical trend shown by the measured core radius $r_{\rm core}$ as a function of $\nu$, we infer that the synchrotron opacity model may not be valid for all cases. A Fourier periodogram analysis yields power law slopes in the range $-1.6$ to $-3.5$ describing the power spectral density shape and gives bend timescales in the range $0.52 - 0.66~$yr. This result, and both positive and negative $\alpha$, indicate that the flares originate from multiple shocks in a small region. Important objectives met in our study include: the demonstration of the computational efficiency and statistical basis of the piecewise Gaussian fit; consistency with previously reported results; evidence for the core shift dependence on observation frequency and its utility in jet diagnostics in the region close to the resolving limit of very long baseline interferometry observations.