If my Kalman Filter ignores measurements for being old, does that make it ageist? But actually, how do you update with old measurements/Should I?

Question

Applying code here https://github.com/JunshengFu/tracking-with-Extended-Kalman-Filter/tree/master/src (code snippet from FusionEKF.cpp) and it looks like there's an assumption of it only being updated forward in time and not backwards. Besides it being annoyingly only in 2D, Q looks wrong for negative dt's.

// compute the time elapsed between the current and previous measurements
  float dt = (measurement_pack.timestamp_ - previous_timestamp_) / 1000000.0;  //  in seconds
  previous_timestamp_ = measurement_pack.timestamp_;

  float dt_2 = dt * dt;
  float dt_3 = dt_2 * dt;
  float dt_4 = dt_3 * dt;

  // Modify the F matrix so that the time is integrated
  ekf_.F_(0, 2) = dt;
  ekf_.F_(1, 3) = dt;

  //set the process covariance matrix Q
  ekf_.Q_ = MatrixXd(4, 4);
  ekf_.Q_ << dt_4/4*noise_ax,   0,                dt_3/2*noise_ax,  0,
             0,                 dt_4/4*noise_ay,  0,                dt_3/2*noise_ay,
             dt_3/2*noise_ax,   0,                dt_2*noise_ax,    0,
             0,                 dt_3/2*noise_ay,  0,                dt_2*noise_ay;

  ekf_.Predict();

  /*****************************************************************************
   *  Update
   ****************************************************************************/

  /**
     * Use the sensor type to perform the update step.
     * Update the state and covariance matrices.
   */

  if (measurement_pack.sensor_type_ == MeasurementPackage::RADAR) {
    // Radar updates

    H_jacobian = tools.CalculateJacobian(ekf_.x_);
    ekf_.H_ = H_jacobian;
    ekf_.R_ = R_radar_;
    ekf_.UpdateEKF(measurement_pack.raw_measurements_);
  }
  else {
    // Laser updates
    ekf_.H_ = H_laser_;
    ekf_.R_ = R_laser_;
    ekf_.Update(measurement_pack.raw_measurements_);
  }