Hands-On With Business Analytics
75 students in 15 teams
Business analytics is the use of data to support business decisions. This module aims to give students a practical, hands-on experience with the tools and techniques used in business analytics, especially dealing with Big Data analytics. The course focuses on the correct interpretation and use of data. We will cover technical definitions of segmentations, correlations, predictions and causation and how it can be used in business settings.
In the United States, Federal Law requires that all employers have to pay workers a minimum wage. Minimum wage policy is used as a policy instrument to help the poor, and over the years, the state average has been increasing. In this project, we study if increases in minimum wage have an unintended consequence of increasing the occurrence of Driving-While-Intoxicated (DWI) accidents. We focus mainly on the younger members of the population, whose disposable incomes are more likely to be affected by changes in minimum wage policy, and who have a higher income elasticity in the purchase of alcohol.
In this project, we will take a deeper look at the casual relationship between weather and public transportation ridership in Singapore from 1st, Jan., 2015 to 31st, Dec., 2016. There’s an intimate relationship between weather conditions and riderships of public transportation, our project will compare weather condition impact to ridership based on daily time period, days of a week, holidays of a year. We suggest the transit agencies of singapore should develop specific policy measures to rearrange the public transport service frequency according to different weather conditions, time period of a day and specific days of a year.
What makes a particular type of content go viral? Why do some articles become an online sensation and why some others don’t? Such questions have gained immense attention from both industry and researchers, particularly in the context of online media. Our project intends to develop a mechanism that will help both content consumers and content producers in a range of data-driven inquiries, such as to comparatively analyze articles and content tags to explain and to predict future popularity, as well as identify articles that have a high propensity to go viral.
Abstract: Consumption of marijuana in U.S is increasing due to legalization of cannabis in numerous states. However, legalization remains controversial because of safety issue, including driving under the influence. It is, indeed, the most frequently found illicit drug in the blood of the drivers of vehicle crashes. In this project, the causality of legalization of recreational usage of cannabis and fatal traffic accidents will be examined using dataset from National Highway Traffic Safety Administration. Counties in non-legalized states adjacent to legalized states will be selected as a treatment group and compared with counties remote from the border with legalized states.
Market capitalisation of the cryptocurrency space increased from $17b in Jan 17 to a peak of $795b in Jan 18. As prices soared faster than the increase in difficulty, an activity called mining, which awarded miners with small amounts of the cryptocurrency in return for computationally intensive tasks, became relatively more profitable. Recent studies have shown that bitcoin mining by itself is using more energy than necessary to power 150 individual countries. Whilst current electricity draw is mitigated in part by geothermal sources, a rise in electricity consumption should theoretically draw upon coal powered electricity with measurable air pollution levels.
On 23rd June 2016, 52% British voters chose to leave the European Union Britain and gain the freedom to manage its own affairs. Brexit Impacted the UK in many ways. One serious issue the country faced was safety. We noticed that the crime reports in some regions of London increased from Apr 2016 to May 2017, and we want to find out whether Brexit is the underlying cause of increase in anti-social crimes. We try to find the causality of Brexit referendum on crime rate using London's crime datasets from 2013 to 2017 and the voting results in the corresponding regions using socioeconomic factors (e.g: Percentage of immigrants in the region, economic status) as the mediating variables.
311 is the City of Houston’s non-emergency department directory and self-service city request line which is designed to make city government more user-friendly and responsive to city residents. Houston citizens are able to request city services and report problems anytime through this government service hotline, from graffiti removal to street hazard, which roughly reflect the needs of the Houston and its residents. While most studies usually try to explore the relationship between city neighbourhoods and economic or social makeup, our study tends to investigate the influence of weather on citizens’ daily needs. Will a heavy rain result in more requests on traffic-related service request? As city service request is also a measurement of city-based dissatisfaction, we can furthermore identify temporal changing pattern of citizens’ tolerance on city imperfections under environmental impact.
Smoking is the leading cause of preventable death. More than 16 million Americans live with a disease by smoking. Besides, smoking costs Americans billions of dollars every year, including direct medical care fee and exposure to second-hand smoking. Understanding whether the policy is effective and consumer cigarette expenditure pattern can better protect public health. Clean indoor air policy is an effective way to eliminate exposure to second-hand smoke. For policymakers from different states, the result of the study can push them to implement this act, in order to reduce public smoking, protect public health, and save massive medical expenditure. The objective of this project is to understand whether statewide clean indoor air act causes the reduction of cigarette expenditure.
Since 911 attack, security has been tightened at the airports and longer time is required for screening before flights. In Dec 2011, Transportation Security Administration (TSA) has introduced Pre-check service at some airports to expedite the screening process. This project aims to find out if the introduction of this program will help to reduce flight delays in the participating airports. This project will work with airports and flight information datasets. Methodology such as Difference-In-Differences(DiD) is used to establish causality between flight delays and the pre-check service.
The project seeks to study causality in determining if increase in ethanol demand driven by renewable fuel standards cause increase in price of food commodities that are biofuel feedstocks. In particular, the project will focus on ethanol production in the U.S and the impact on its primary feedstock, corn and other food crops.
Young Drivers(15-24 years old) make up 10.9% of total UK driving licence holders. Drivers aged 21-24 make up 4.7% of total driving license, but are involved in most of fatal and serious crashes. However, there is a rising concerns about road accidents caused by young drivers because of their relative inexperience and impulsive behaviors. Does the age of the motorbike rider influence the number of traffic accidents and the casualty ratio? In the year 2013, the UK government issued a new policy to increase the minimum age of motorbike rider from 21 to 24 based on european requirements. This study will use the UK traffic accident dataset to do causal analysis about the effect of the new policy. Difference in difference(DID) model and propensity score matching(PSM) will be implemented to isolate the treatment effect of the new policy.
Tipping culture has been unique to restaurants in the United States. It is often perceived to be a reward for hardworking servers who have provided good service. However, due to low wages of these workers in the service sector, our group feel that tipping have instead transformed into a social norm in the United States. Hence, in this project, our group aims to study the effects of tipping on customer satisfaction. This is done through analyzing restaurants who have implemented no-tipping policies in recent years, and the effects on its customers.
There are certainly lots of mysteries surrounding the Moon, Earth's sole natural satellite. From ancient Chinese superstitions, to sci-fi phenomena such as werewolf transformations. All of these make lunar phases both intrinsically interesting, and potentially useful when correlated with other types of data. One of which is human behavior: the term "lunatic", an informal term for a person with unsound behaviour, was derived on early beliefs tagging several mental diseases to be caused by the moon. In the U.K., a research carried out by the Sussex Police concluded that there is a strong correlation between reported violent incidents and the occurence of full moons. This prompted the police department to deploy more police patrollers during full moon nights. In this project, we'll take a deeper look into this alleged correlation. By developing our own model, we'll try to verify if full moon occurences and crime rates are directly correlated, or if it is affected by other variables and finally debunk this myth. At a statistical point of view, did the Sussex Police make a sound decision? Should we also be deploying more police force during full moon nights? Stay tuned with us to find out.
This study is designed to test whether there is causal effect between kids indulgence in music and their knowledge in mathematics. Using panel data from United States, a fixed effect model will be used to determine whether kids who are more actively involved with music and using any kind of musical instruments have shown significantly better performance in mathematical problem-solving skills. In addition to this, other important factors that will be used to predict the relation include parental influence, educational environment, health, tuition etc.
The number of over-70s who hold a UK driving licence had exceeded four million for the first time in 2013, causing concerns on road safety. This study examined the causal relation between the involvement of elderly drivers in traffic accidents and the injury severity of casualties. The 2005 – 2015 UK Car Accident Data, together with historical weather condition data crawled from Weather Underground, was used to build fixed effect model on the proportion of elderly drivers per year at county level against the average casualty severity. Instrument variables were also applied to eliminate the potential endogeneity issue.