Personal statement 2

Personal Statement
I have always been fascinated by statistics right from a very young age. I developed mathematical thinking from my sophomore year which has been very useful in real life. My desire to pursue the Ph.D. program in statistics at CMU (Carnegie Mellon University) has been motivated by the wide knowledge and experience gained through providing solutions to commercial, environmental, medical and other problems. I decided to devote my life doing research in statistics. My career path has also been shaped by the passion, my previous studies, research experiences and my commitment towards achieving my dream. I am eager to attain more knowledge in statistics to improve my capability in research. I envision the Ph.D. program in Statistics as an ideal platform to further extend my knowledge. I believe that this program is likely to improve my merits and competence in my area of specialization.
I have laid a firm foundation in statistics as well as computer science through my previous studies and research experiences. My academic background has been a great pillar and a stepping stone that has enabled me to make significant achievements in mathematics and statistics. I opted for mathematics as my bachelor degree, and after strict selection, I was admitted to Hua Loo-Keng Class, the only top class in college of mathematics, jointly educated by Chinese Academy of Sciences and Shandong University. I scored highly in mathematics courses, especially in mathematical statistics, probability theory and stochastic process.

Besides the degree program, I also took many courses such as C, C++, MATLAB, SQL and data structure to strengthen my statistical computing ability. In the course of my final year, I was selected to join Chinese Academy of Sciences for a one-year study where I took advantage of graduate courses such as real analysis and partial differential equation.

My journey of statistics was spurred by various undergraduate research training programs that seemed to me as my first step to the real world of statistics. I was able to learn about algorithms such as neural network, random forest and support vector machine, and developed my programming skills about R and MATLAB to process and cleanse extremely large data volumes. Attracted by statistical principle behind recommender system, our project aimed at predicting online users’ consumption behaviors through learning their previous transaction records and browsing history. I attended seminars about machine learning and data mining organized by a professor in the school of computer science and technology to acquire a deep understanding of machine learning. Through active participation, I led the team in applying the gradient boosting decision tree algorithm on a public database provided by Tmall Shopping Website and finally obtained a 0.375 F-score, exceeding many other works. The project greatly strengthened and polished my leadership and communication skills.
I later joined Professor Hui Huang’s group in the center for statistical science, Peking University during the summer vacation to acquire a deeper knowledge of the link between machine learning and traditional statistics methods and their application. We worked on a temporal and spatial statistical model for air pollutants, a subproject of a national key scientific research project led by Professor Songxi Chen. I was able to explore time series model, conditional random field, and neural network but none of them alone fit pollutant data very well. However, after reading a paper about influenza tracking model that combines auto-regression and hidden Markov models, I applied its R package and adjusted it to our data. The result improved better. The project inspired me since it was an eye opener that not every problem could be easily solved by just applying some black-box machine learning algorithm especially those without sufficient data or with time-related data that may cause concept drift. We need to carry out analysis of the intrinsic characteristic for any given set of data to pick the best approach.
Through my journey of statistics, I also developed interests in unsupervised learning problems. I was attracted by the ability of wavelet neural network (WNN) to learn better local nonlinear continuous and fast changing function. In a group of three students, we employed it to clustering. We were able to utilize the particle swarm optimization (PSO) algorithm to train our wavelet neural network to overcome low efficiency of gradient descent method and avoid local minimum. During the process, I was charged with the responsibility of conducting a literature review on unsupervised neural networks and other clustering approaches. I also tested our clustering algorithm by applying it to image segmentation and did an experiment to compare it with conventional KNN clustering algorithms. Our results showed a higher precision of segmentation as well as a faster training speed compared with conventional techniques. I am motivated to mention that our paper has been accepted for publication in IEEE Xplore and Oral presentation at “IEEE International Conference on Communication and Electronics Systems” (IEEE ICCES 2016).
Getting an opportunity to pursue the Ph.D. program in statistics will be a relief that my dream will finally come true. Based on my future career plan, I wish to dedicate myself to the research in the field of statistics. I am confident that I will acquire advanced methods of data analysis and abundant experience about addressing crucial cross-disciplinary questions in CMU and become an excellent and competent researcher. Equipped with a vast knowledge of statistics, I will be able to help decision makers make informed choices. Based on my knowledge and experience acquired, I have been keen in dedicating myself to the area of statistical research. Upon being granted the opportunity to pursue the program, I will have the pleasure of taking the challenge to accomplish the research. I plan to join research organizations both in my country and abroad to help bring solutions to the real problems. I also plan to share the knowledge by teaching statistics at the University level. The Ph.D. program in statistics at your institution is extremely relevant to my future career plans.
My desire to pursue the Ph.D. program in statistics at CMU (Carnegie Mellon University) is in line with my professional experience hence it is an indispensable opportunity. I believe it will provide immeasurable knowledge to mold my mathematical thinking ability by improving my merits and competence in my area of specialization. Based on my research experience, previous studies, my passion, and commitment, I am convinced that I am ready for a change and the best place to spur this change is at your institution. I have devoted my life doing research in statistics, and I am extremely hopeful to become one of the best professional experts in statistics. I am optimistic that I will be considered for the admission in your institution.