We’re often warned that the internet will hasten the dating apocalypse. The internet (it is posited) is depriving us of the elusive in-person magic, and modern courtship is now little more than love at first byte. There remains uncertainty, however, about what the independent impact of the internet on the dating market has been. Similar to the internet, the telephone also changed the way we communicate, but its effect on the dating market was mostly complementary to the 'traditional' ways of meeting – i.e. calling your school crush at home. So the question remains: Is the effect of the internet on the dating market complementary (adding your school crush on Facebook) or substitutionary (matching with a stranger on Tinder)? Is the internet any better than the telephone? If all that was known about a random couple is that they met after 2015, I find that there is a 1 in 3 chance that the couple met as strangers online. Lesbian couples who met after 2015 have a 1 in 2 chance of meeting online, whereas gay male couples have a 63% probability of meeting online as strangers. This increased likelihood of same-sex couples meeting online (as opposed to heterosexual couples) confirms the thin-market hypothesis. The key value proposition of the internet is that it reduces search frictions in the dating market – effectively making it easier for individuals to seek out their optimal matching. I find that the internet is primarily displacing only ‘social circles’ as a dating venue – the probability of meeting partners in public or at institutions (like college) is unchanged. In other words – individuals are essentially replacing their friends with Wi-Fi when it comes to mate search.

預測市場的研究近年來在學界逐漸受到重視，因為它利用價格具有訊息加總的功能，每每創造出良好的預測績效，但一個預測市場的建立在諸多原因下，通常不易吸引大規模的參與者，例如為免觸犯法令規定，以虛擬貨幣代替真錢進行交易，在缺乏真實貨幣的獲利誘因下，很難有效吸引參與者，即便真能以真實貨幣交易，若實驗的議題並非一般大眾感興趣的話題也不易吸引多數人參與，在這種情況下無法避免要面臨市場交易過於淺薄的問題，雖然不少文獻標榜淺薄市場不會影響預測市場的預測精準度，但並不表這是一個可以置之不理的問題。 本文以預測市場預測2006年北高市長選舉為實證基礎，闡明淺薄市場對價格產生的影響，以及這些影響將導致對未來事件的錯誤評價與推論，要避免這種錯誤的評價與推論唯有設法消除淺薄市場引發的干擾，因此我們提出了五種可以消除這些干擾的方法並從中選擇一較佳者。如同一般文獻的讚揚，我們再次從預測市場獲得精確的預測效果，同時證明所謂淺薄市場不影響預測市場的預測精準度前提乃在消除淺薄市場對價格產生的干擾之後才能還原這個真相。