def report(self): if self.valid_confidence_interval(): print "Intervalos de confiança estimados válidos!, exibindo os resultados:" print "E[N1]: ", est.mean(self.results['m_s_N1'], self.samples) print "IC - E[N1]", est.confidence_interval(self.results['m_s_N1'], self.results['m_s_s_N1'], self.samples) print "E[N2]: ", est.mean(self.results['m_s_N2'], self.samples) print "IC - E[N2]", est.confidence_interval(self.results['m_s_N2'], self.results['m_s_s_N2'], self.samples) print "E[T1]: ", est.mean(self.results['m_s_T1'], self.samples) print "IC - E[T1]", est.confidence_interval(self.results['m_s_T1'], self.results['m_s_s_T1'], self.samples) print "E[T2]: ", est.mean(self.results['m_s_T2'], self.samples) print "IC - E[T2]", est.confidence_interval(self.results['m_s_T2'], self.results['m_s_s_T2'], self.samples) print "E[Nq1]: ", est.mean(self.results['m_s_Nq1'], self.samples) print "IC - E[Nq1]", est.confidence_interval(self.results['m_s_Nq1'], self.results['m_s_s_Nq1'], self.samples) print "E[Nq2]: ", est.mean(self.results['m_s_Nq2'], self.samples) print "IC - E[Nq2]", est.confidence_interval(self.results['m_s_Nq2'], self.results['m_s_s_Nq2'], self.samples) print "E[W1]: ", est.mean(self.results['m_s_W1'], self.samples) print "IC - E[W1]", est.confidence_interval(self.results['m_s_W1'], self.results['m_s_s_W1'], self.samples) print "E[W2]: ", est.mean(self.results['m_s_W2'], self.samples) print "IC - E[W2]", est.confidence_interval(self.results['m_s_W2'], self.results['m_s_s_W2'], self.samples) print "V(W1): ", est.mean(self.results['v_s_W1'], self.samples) print "IC - V(W1)", est.confidence_interval(self.results['v_s_W1'], self.results['v_s_s_W1'], self.samples) print "V(W2): ", est.mean(self.results['v_s_W2'], self.samples) print "IC - V(W2)", est.confidence_interval(self.results['v_s_W2'], self.results['v_s_s_W2'], self.samples) else: print "Intervalos de confiança estimados inválidos. Aumente o numero de amostras ou de clientes por amostra"
def valid_confidence_interval(self): return (2.0*est.confidence_interval(self.results['m_s_N1'], self.results['m_s_s_N1'], self.samples) <= 0.1*est.mean(self.results['m_s_N1'], self.samples)) and \ (2.0*est.confidence_interval(self.results['m_s_N2'], self.results['m_s_s_N2'], self.samples) <= 0.1*est.mean(self.results['m_s_N2'], self.samples)) and \ (2.0*est.confidence_interval(self.results['m_s_T1'], self.results['m_s_s_T1'], self.samples) <= 0.1*est.mean(self.results['m_s_T1'], self.samples)) and \ (2.0*est.confidence_interval(self.results['m_s_T2'], self.results['m_s_s_T2'], self.samples) <= 0.1*est.mean(self.results['m_s_T2'], self.samples)) and \ (2.0*est.confidence_interval(self.results['m_s_Nq1'], self.results['m_s_s_Nq1'], self.samples) <= 0.1*est.mean(self.results['m_s_Nq1'], self.samples)) and \ (2.0*est.confidence_interval(self.results['m_s_Nq2'], self.results['m_s_s_Nq2'], self.samples) <= 0.1*est.mean(self.results['m_s_Nq2'], self.samples)) and \ (2.0*est.confidence_interval(self.results['m_s_W1'], self.results['m_s_s_W1'], self.samples) <= 0.1*est.mean(self.results['m_s_W1'], self.samples)) and \ (2.0*est.confidence_interval(self.results['m_s_W2'], self.results['m_s_s_W2'], self.samples) <= 0.1*est.mean(self.results['m_s_W2'], self.samples)) and \ (2.0*est.confidence_interval(self.results['v_s_W1'], self.results['v_s_s_W1'], self.samples) <= 0.1*est.mean(self.results['v_s_W1'], self.samples)) and \ (2.0*est.confidence_interval(self.results['v_s_W2'], self.results['v_s_s_W2'], self.samples) <= 0.1*est.mean(self.results['v_s_W2'], self.samples))
def calc_results(self): self.results = { 'E[N1]' : { 'value' : est.mean(self.sums['m_s_N1'], self.samples), 'c_i' : est.confidence_interval(self.sums['m_s_N1'], self.sums['m_s_s_N1'], self.samples) }, 'E[N2]' : { 'value' : est.mean(self.sums['m_s_N2'], self.samples), 'c_i' : est.confidence_interval(self.sums['m_s_N2'], self.sums['m_s_s_N2'], self.samples) }, 'E[T1]' : { 'value' : est.mean(self.sums['m_s_T1'], self.samples), 'c_i' : est.confidence_interval(self.sums['m_s_T1'], self.sums['m_s_s_T1'], self.samples) }, 'E[T2]' : { 'value' : est.mean(self.sums['m_s_T2'], self.samples), 'c_i' : est.confidence_interval(self.sums['m_s_T2'], self.sums['m_s_s_T2'], self.samples) }, 'E[Nq1]' : { 'value' : est.mean(self.sums['m_s_Nq1'], self.samples), 'c_i' : est.confidence_interval(self.sums['m_s_Nq1'], self.sums['m_s_s_Nq1'], self.samples) }, 'E[Nq2]' : { 'value' : est.mean(self.sums['m_s_Nq2'], self.samples), 'c_i' : est.confidence_interval(self.sums['m_s_Nq2'], self.sums['m_s_s_Nq2'], self.samples) }, 'E[W1]' : { 'value' : est.mean(self.sums['m_s_W1'], self.samples), 'c_i' : est.confidence_interval(self.sums['m_s_W1'], self.sums['m_s_s_W1'], self.samples) }, 'E[W2]' : { 'value' : est.mean(self.sums['m_s_W2'], self.samples), 'c_i' : est.confidence_interval(self.sums['m_s_W2'], self.sums['m_s_s_W2'], self.samples) }, 'V(W1)' : { 'value' : est.mean(self.sums['v_s_W1'], self.samples), 'c_i' : est.confidence_interval(self.sums['v_s_W1'], self.sums['v_s_s_W1'], self.samples) }, 'V(W2)' : { 'value' : est.mean(self.sums['v_s_W2'], self.samples), 'c_i' : est.confidence_interval(self.sums['v_s_W2'], self.sums['v_s_s_W2'], self.samples) } }