All Figures

$\begin{figure} \par\includegraphics[width=12.4cm,clip]{gausscomp.ps} \end{figure}$ Figure 1: This figure shows the comparison between the distribution of the observations used to generate the virtual light curves (dashed line) and a Gaussian distribution (solid line). Notice the excess where the normalized distance from the center of the distribution is greater than 1.1. Despite the fact that it looks large, it is actually due to just 10 out of the 10⁴ observations.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=12cm,clip]{rocs.ps} \end{figure}$ Figure 2: The plot on the left shows the Receiver Operating Curves for all five detectors used in the simulations for the fixed set of parameters $(N_{\rm tr} = 3, r_{{\rm obs}} = 5, d = 0.5)$ . The curved solid line is the matched filter, the long dashed line is the correlation, the dotted line is Deeg's approach, the short dashed line is the BLS and the dot-dashed line is the Bayesian. The straight solid line shows the limit that corresponds to random guessing. The plot on the right shows the same diagram for the matched filter approach and different transit parameters. In the format $(N_{\rm tr}, r_{{\rm obs}},d)$ , the curved solid line is (6,5,0.5), the long dashed line is (3,5,0.5), the dotted line is (6,5,0.25), the short dashed line is (3,5,0.25) and the dot-dashed line is (3,5,0.1). The lines from the other sets of input parameters tested would be nearly indiscernibly crowded into the upper left-hand corner of the diagram, as they represent stronger transit signals.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=12.3cm,clip]{Pdstraight.ps} \end{figure}$ Figure 3: A plot of $\ln(1-(\ln(1-P_{{\rm D}}))$ against $N_{{\rm obs}} \times \sigma ^{2}$ , where $P_{{\rm D}}$ is the probability of detection for a false alarm rate of 10^-2, $N_{{\rm obs}}$ is the number of observations while the planet is in transit and $\sigma$ is the depth of the transit observed. The values in this plot are taken from the matched filter approach column in Table 1. The point corresponding to (10,10,0.5) is missing because, due to the limited number of events in the simulation, its $P_{{\rm D}}$ equals one and is therefore undefined in the function used. The asterisks show the actual values from Table 1, while the dashed line (with a slope of 0.1668) is the best fit straight line to these values.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=12.4cm,clip]{gausscomp.ps} \end{figure}$	Figure 1: This figure shows the comparison between the distribution of the observations used to generate the virtual light curves (dashed line) and a Gaussian distribution (solid line). Notice the excess where the normalized distance from the center of the distribution is greater than 1.1. Despite the fact that it looks large, it is actually due to just 10 out of the 10⁴ observations.
Open with DEXTER