Update paper.md

one citep to remove and put citet

paper/paper.md

@@ -55,7 +55,7 @@ Pyinterpolate performs six types of spatial interpolation; inverse distance weig
 4. **Area-to-Area Poisson Kriging** is used for areal interpolation and filtering. The point-support allows the algorithm to filter unreliable rates and makes final areal representation of rates smoother.
 5. **Area-to-Point Poisson Kriging** where areal support is deconvoluted in regards to the point support. Output map has a spatial resolution of the point support while coherence of analysis is preserved (sum of rates is equal to the output of Area-to-Area Poisson Kriging). It is used for point-support interpolation and data filtering.
 
-The theory of Kriging is described in supplementary materials in the [paper repository](https://github.com/SimonMolinsky/pyinterpolate-paper/blob/main/paper/supplementary%20materials/theory_of_kriging.md) or in more detail in @Armstrong:1998. [@OliverWebster:2015] point to the practical aspects of Kriging. The procedure of the interpolation with Poisson Kriging is presented in @Goovaerts:2006 and the semivariogram regularization process is described in @Goovaerts:2007.
+The theory of Kriging is described in supplementary materials in the [paper repository](https://github.com/SimonMolinsky/pyinterpolate-paper/blob/main/paper/supplementary%20materials/theory_of_kriging.md) or in more detail in @Armstrong:1998. @OliverWebster:2015 point to the practical aspects of Kriging. The procedure of the interpolation with Poisson Kriging is presented in @Goovaerts:2006 and the semivariogram regularization process is described in @Goovaerts:2007.
 
 The comparison to existing software is presented in the supplementary document [here](https://github.com/SimonMolinsky/pyinterpolate-paper/blob/main/paper/supplementary%20materials/comparison_to_gstat.md), Ordinary Kriging outcomes are compared for *gstat* and Pyinterpolate.