axs[0, 1].plot(true_x_sig, np.repeat(50, true_x_sig.size), color='k', lw=2)
axs[1, 1].plot(phadke_x_sig,
np.repeat(50, phadke_x_sig.size),
color='k',
lw=2)
axs[2, 1].plot(isb_x_sig, np.repeat(50, isb_x_sig.size), color='k', lw=2)
axs[3, 1].plot(isb_norm_x_sig,
np.repeat(50, isb_norm_x_sig.size),
color='k',
lw=2)
print('ANOVA')
print('True Axial')
print(extract_sig(spm_one_way_rm_true, x))
print('P-values: ')
print(output_spm_p(spm_one_way_rm_true))
print('ISB')
print(extract_sig(spm_one_way_rm_isb, x))
print('P-values: ')
print(output_spm_p(spm_one_way_rm_isb))
print('ISB Norm')
print(extract_sig(spm_one_way_rm_isb_norm, x))
print('P-values: ')
print(output_spm_p(spm_one_way_rm_isb_norm))
print('Phadke')
print(extract_sig(spm_one_way_rm_phadke, x))
print('P-values: ')
print(output_spm_p(spm_one_way_rm_phadke))
print('ANOVA Post-hoc')
p_critical = spm1d.util.p_critical_bonf(alpha, 3)
np.repeat(spm_y[idx_act, 1], isb_x_sig.size),
color=color_map.colors[0],
lw=2)
axs[idx_act, 1].plot(isb_norm_x_sig,
np.repeat(spm_y[idx_act, 1] - 3,
isb_norm_x_sig.size),
color=color_map.colors[3],
lw=2)
# print significance
print('Activity: {} Joint: {}'.format(activity, traj_name.upper()))
print('ISB vs True')
print(extract_sig(isb_vs_true, x))
print('Max: {:.2f}'.format(np.abs(isb_mean[-1] - true_mean[-1])))
print('P-values: ')
print(output_spm_p(isb_vs_true))
print('ISB Rectified vs True')
print(extract_sig(isb_norm_vs_true, x))
print('Max: {:.2f}'.format(np.abs(isb_norm_mean[-1] - true_mean[-1])))
print('P-values: ')
print(output_spm_p(isb_norm_vs_true))
print('Phadke vs True')
print(extract_sig(phadke_vs_true, x))
print('Max: {:.2f}'.format(np.abs(phadke_mean[-1] - true_mean[-1])))
print('P-values: ')
print(output_spm_p(phadke_vs_true))
if idx_act == 0:
# legend lines
mean_left_lns.append(true_left_ln[0])
mean_left_lns.append(phadke_ln[0])